Skip to main content

Full text of "Inorganic seminar abstracts"

See other formats


UNIVERSITY OF 

ILLINOIS LIBRARY 

AT URBANA-CHAMPAIGN 



Digitized by the Internet Archive 

in 2012 with funding from 

University of Illinois Urbana-Champaign 



http://archive.org/details/inorganicsemi73747879univ 



TJ CO DENTS 



1/ 



INORG 

S umc r 3 e s s i on : P /: 

IRUPOLE RESONANCE STU OF COBALT (ill) COMPLEXES - 1 
Robert A. LaRossa 

CHEMISTRY AND BONDING IN STJBST] ID COBALT CARBONYL COMPLEXES - 2 
Thomas E. Boyd 

KEGAND EXCHANGE STUDIES OF METHYLATOCOBALT(lIl) COMPLEXES - ^ 

Randolph J. Guschl 

STRUCTURAL AND BONDING PROPERTIES OF SMALL RING COMPOUNDS -V* ubrary ai the 5 

Yu Wang 

OCT Z\ 1974 
BONDING AND STEREOCHEMICAL STUDIES OF GROUP IA AND IIA 6 

ORGANOMETALLIC COMPOUNDS - Richard Zerger HffiScilaSB 

NUCLEAR MAGNETIC RESONANCE STUDIES OF ORGANOMETALLIC EXCHANGE 9 
REACTIONS - Richard L. Kieft 

Fall Session : 

A MOLECULAR ORBITAL DESCRIPTION OF LEWIS ACID-BASE INTERACTIONS - 10 
Allen Marks 

THE REACTIVITY AND MAGNET OCHEMISTRY CF SELECTED METALLOCENES - 11 
Mary Ellen Switzer 

SPECTROSCOPIC STUDIES OF PENT' A- COORDINATE ORGANOMETALLIC 14 

COMPLEXES - Cheryl D. Pribula 

.KINETICS AND MECHANISMS OF THE SUBSTITUTION REACTION OF l6 

ALKYL-GROUP IV MEDAL COBALT -TETRACARBONYLS - James Eaton 

ELECTRON TRANSFER REACTIONS OF COBALT COMPLEXES - 17 

Michael Broccardo 

OXIDATIVE DEHYDROGENATION AND KINETICS OF RUTHENIUM ( II )- 20 

ETHYLENEDIAMINE COMPLEXES - Dennis Mahoney 

BIOCHEMICAL APPLICATIONS OF LANTHANIDE SHIFT REAGENTS - 22 

Pamela A. Milton 

NUCLEAR RELAXATION PROCESSE! : APPLICATIONS - Dennis R. Kidd 25 

MANGANESE PORPHYRINS AND PHTHALOCYANINES - Lynne M. Parr 23 

THE ACTIVE SITE OF ME] T1 rHTN - Mamoru Tachikawa 30 

HOMOGENEOUS INORGANIC OSCILLATING REACTIONS - Ben Fieselmann 33 



-2- 

THE SPIN LABEL TECHNIQUE AND ITS SSI: . : LICATIONS - 36 

Marinda P. Li 

OBSERVING THE BEHAVIOR OP TRANSIENT CHEMICAL SPECIES - 39 

Mitch Hoselton 

SECONDARY PROCESSES IN PHOTOELECTRON SPECTROSCOPY - 4l 

Edward J. Laskowski 

magnetic exchange TrrrET.^erioNS *:.« dpisric cu(ii) complexes - 45 

E] vira F. Hasty 

BINUCLEAR COMPLEXES OF COBALT (il) - Nelson B. 0' Bryan 47 

BONDING IN TRANSITION METAL BINARY HYDRIDES - 49 

Ma'mun Absi-Halabi 

CALORIMETRIC STUDIES OF LEWIS ACID-BASE INTERACTIONS - 53 

R. Martin Guidry 

ELECTRONIC INTERACTIONS BETWEEN PAIRS OF METAL IONS 56 

BRIDGED BY SMALL ANIONS - D. Michael Duggan 

POLY(l-PYRAZOLYL) BORATE COMPLEXES AND ORGANOMETALLICS - 58 

Robert Mink 

HYDRIDE AND ALEYL DERIVATIVES OF NICKEL - Alex N. Williamson 6l 



-1- 



Nuclear Quadrupole Resonance Studies of Cobalt (ill) Complexes 

Robert A. LaRossa July 2, 1973 

One of the most interesting outgrowths of research on the proper- 
ties of Vitamin Bi 2 and its derivatives resulted from the observation 
that biochemically important forms, called coenzyme Bi 2 and methyl - 
cobalamin, exist in which there are direct cobalt-carbon bonds. This 
prompted the synthesis of a number of simple cobalt (ill) complexes 
containing alkyl groups directly bound to cobalt. Among the most ex- 
tensively studied molecules of this type are the Lewis base adducts of 
methylatobis(dimethylglyoximato) cobalt (ill) ( "cobaloximes" ) . 

Since nuclear quadrupole resonance (nqr) spectroscopy is a sensi- 
tive indicator of the electron distribution around a quadrupolar nucleus, 
a study of 59 Co and 35 C1 nqr in cobaloximes and related complexes should 
provide information pertinent to the question of bonding in these com- 
pounds. Prior to this study, 59 Co nqr had been observed in only a 
limited number of six-coordinate Co (ill) compounds. The observed trans- 
ition frequencies were generally below 12 MHz. Therefore, a nuclear 
quadrupole resonance spectrometer of the superregenerative type, which 
had optimum sensitivity in this low frequency region, was developed and 
used to detect most of the resonances reported in this seminar. 59 Co 
and 35 C1 quadrupole coupling constants and 59 Co asymmetry parameters 
were determined for a series of compounds XCo(dh) 2 L, where dh = dimethyl- 
glyoxime monoanion, X = CH 3 , CI , Br , CHCI2, L = Lewis base. 59 Co 
chemical shifts were also measured for most of the compounds in this 
series. 

A description of the generation and orientation of the 59 Co field 
gradient tensor in terms of d-orbital populations is only partially 
successful. However, by using a "donated charge" model and taking the 
geometry of the complexes into account, it is possible to determine how 
the quadrupole coupling constant and asymmetry parameter should change 
as the interactions with the ligands X and L change. Taken in con- 
junction with the 59 Co nmr shifts, the measured nqr parameters may be 
interpreted in terms of this model. 



2 



CHEMISTRY AND BONDING IN SUBSTITUTED 
COBALT CARBONYL COMPLEXES 

Thomas E. Boyd August 17, 1973 

The study of metal-metal bonding among the transition metals 
is a relatively new development. In this research, three types 
of compounds were Investigated: (l) dimeric compounds of the form 
[Co(C0J3P] 2 ; (2) heteronuclear metal-metal bonded compounds of the 
type R 3 MCo(C0)3P; and (3) compounds having a higher degree of 
phosphorus base substitution. In these systems, M represents a 
group IV metal, R is an organic group, and P denotes a phosphine 
or phosphite. 

Previous work has established that the dimers, [Co(C0)3P] 2 , 
consist of two axially substituted tetracarbonyl cobalt moieties 
joined by a cobalt-cobalt bond. This is not true of the P(0CH2) 3 CC 2 H 5 
derivative, however, in which the phosphorus ligands occupy radial 
positions. The electronic spectra of these compounds were investi- 
gated in oriented nematic phase to determine the polarization of 
absorptions observed between 350-400 nm. These appear to be polar- 
ized along the cobalt-cobalt bond which is consistent with their 
assignment as o-*o* transitions. These studies showed that cobalt- 
cobalt bond strength is dictated by the effective a-donor strength 
of the phosphorus ligands. From the 9 Co nqr spectra, the nuclear 
quadrupole coupling constant, eQq, was found to increase as the 
Lewis basicity of the phosphorus ligands decreased. In terms of the 
populations of the cobalt 3d orbitals as expressed by the equation 

q 22 = q 320 [N d +i(N d + N d ) - (H d + N d ) ] 
z 2 xz yz xy x 2 -y 2 

in which q is the electric field gradient (efg) at cobalt and 
q 32 o is the efg arising from a 3d 2 electron, a decrease in Lewis 
basicity decreases the N, term causing a lowering of q and thus 

Q y 2 Z Z 

eQq. This is a valid conclusion, as previous work has shown the 
(N d + N. ) term to be larger than the sum of the other two 

xy x 2 -y 2 
terms. The insertion of SnCl 2 into the cobalt-cobalt bond at 6o°C 
exhibited a dependence of reaction rate upon bond strength. How- 
ever, the exact nature of this dependence is uncertain as the 
reaction mechanism has not been elucidated. 

In the investigation of the heteronuclear metal-metal bonded 
compounds, the group IV metal, the group IV metal substituents, and 
the phosphorus base were varied to assess the effects of these 
changes on the metal-cobalt bond. These compounds have a trigonal 
bipyramidal structure with the group IV metal moiety and the 



phosphorus ligand occupying the axial positions. Spectroscopic 
changes resulting from the variation of the group IV metal were 
too small to provide for rigorous interpretation. However, 
when the substituents of these metals were varied, the electronic 
spectra showed a dramatic increase in metal-cobalt bond strength 
as the electronegativity of these groups decreased. The elec- 
tronic, infrared, and nqr spectra shewed a lesser degree of change 
when the phosphorus ligand was varied. As the rr-acceptor strength 
of the phosphorus ligand increases, there is an observed increase 
in metal- cobalt bond strength. Distortion of the three- fold sym- 
metry was noted in the nqr and infrared spectra when the axial 
moieties were substituted with bulky substituents such as phenyl, 
for example. 

Compounds of the type (C s H 5> )3SnCo(C0)4_ x P x in which x < 2 
were investigated. The electronic spectra revealed that the tin- 
cobalt bond strength decreases upon substitution of one carbonyl 
group by a phosphorus ligand, but increases when a second one is 
introduced. Among the disubstituted compounds, two isomers are 
possible. The fact that there was only one set of nqr resonances 
and no observed broadening or splitting in the infrared spectra 
support the notion that one isomer is present. Based on our 
knowledge of the monosubstituted compounds, the disubstituted 
species have a phosphorus ligand in the axial position and one of 
the radial positions. Attempts to prepare compounds having a 
higher degree of phosphorus ligand substitution were unsuccessful. 



-4- 



LIGAND EXCHANGE STUDIES OF METHYliATOCOBALT(lIl) COMPLEXES 

Randolph J. Guschl August 24, 1973 

Variable temperature l9 F and 1 H nmr studies have been obtained 
to determine the rates of axial base exchange from complexes of the 
type CH 3 Co(chel)B, where chel represents a planar tetradentate che- 
late and B is a donor base. For the base l-(2-trifluoromethyl- 
phenyl) imidazole (lm-CF 3 ), nmr spectra in the region of intermediate 
exchange were observed and matched with computer-generated lineshapes 
for chel = N,N ! -ethyl enebis(acetylacetoneiminato) (bae), bisdimethyl- 
glyoximato ((dh) 2 ), diacetylmonoximeimino-diacetylmonoximatoimino- 
propane-1,3 (tmed), 2,3,9,10-tetramethyl-l,4,8,ll-tetraazacyclotetra- 
deca-l,3,8,10-tetraene (tim), and the ring system found in the hepta- 
methyl ester of cobyrinic acid (cobester). From the calculated line- 
shapes, first-order rate constants for dissociation were evaluated and 
comparative AG* values were computed. The order of base lability re- 
flects a cis effect of electron donation by the chelate in the order 
bae > cobester > (dh) 2 > tmed > tim. 

Base adducts of the alky lcobalt (ill) complexes of (dh) 2 , tmed, 
and tim were prepared to observe the effect of replacing the -0H0- 
groups of (dh) 2 with -(CH 2 )3- moieties. Interactions of these systems 
with THF, CH 3 0H, S(CH 3 ) 2 , N(CH 3 ) 3 , 3-fluoropyridine, Im-CF 3 , P(OCH 3 ) 3 , 
and acetonitrile in nitrobenzene and methylene chloride were observed 
and the relative ordering of dissociation rates was found to depend 
on the nature of both the base and the chelate. For any one base, 
exchange is fastest when the chelate is (dh) 2 and slowest with tim. 
This is discussed in terms of changes in the charge at the cobalt 
center and the subsequent changes in a and tt- interact ions. 

The rate of dissociation of P(OCH 3 ) 3 from CH 3 Co(dh) 2 P(0CH 3 ) 3 
was observed in solvents of varying dielectric constant, and the rate 
at 100°C was found to decrease by a factor of about 2 over the range 
of solvents from toluene to nitrobenzene. Replacement of protons in 
the -OHO- bridges of the planar ligand system by deuterium results in 
no observable change in P(OCH 3 ) 3 dissociation rate. A limiting S N 1 
mechanism for dissociation has been proposed involving a transition 
state which is less polar than the ground state, and which does not 
differ much from the ground state in geometry. 

The kinetics data are discussed in terms of a strong cis effect 
of the chelate, and the biological implications of this study are ex- 
plored. The rate of exchange of Im-CF 3 from the Im-CF 3 adduct of 
methylatocobester was found to be 30.0 sec" 1 at 49°C. The result 
implies that axial base exchanges in all alkylcobalt(lll) corrinoids 
are rapid enough to allow base exchange during or before the Co-C 
cleavage characteristic of mechanisms involving these systems. 



-5- 



STRUCTURAL AND BONDING PROPERTIES OP SMALL RING COMPOUNDS 
Yu Wang August 20, 1973 

A problem which has excited a great deal of interest during the 
past several years has been the chemistry and bonding of three and 
four center systems. We have studied the bonding charge distributions 
and structural properties in three- and four-membered ring systems 
containing the second row elements: A three-membered ring compound, 
tetracyanocyclopropane (TCCP), was studied both by X-ray diffraction 
and photoelectron spectroscopy. In this compound, half of the C-C=N 
groups were found to be non-linear and the (CN) 2 C-C(CN) 2 fragment was 
found to be non-planar. A comparison of this compound with transition 
metal olefin complexes 1 will be made. The charge distribution in the 
molecule calculated from photoelectron spectroscopic data showed that 
there is a positive charge region in 

C c 

"c - c^ 
c" ^0 
which is susceptible to nucleophilic attack. 

A four-membered ring compound, squaric acid, was studied both by 
X-ray and neutron diffraction. Assuming D 4 ^ symmetry, the aromatic 
oxocarbon dianion of squaric acid, C 4 Q 4 ? "", has an unusually large 
amount of stabilization energy. 2 The squaric acid structure is planar 
and forms an infinite sheet by means of strong inter-molecular hydrogen 
bonding. Due to the tetragonal pseudo symmetry, the solution of the 
structure proved to be difficult. With the assistance of the single 
crystal infrared and Raman data, 3 the space group was determined to be 
P2x/m with two molecules per unit cell. The molecular geometry dis- 
plays neither D 4 ^ symmetry nor a discrete 1,2-dione structure. Rather, 
the n-electrons are extensively delocalized through the solid by means 
of hydrogen bonding. Thus, shifting the positions of the hydrogen 
atoms results in the formation of tautomeric isomers. In order to 
understand more about the squaric acid aggregate and the extensive tt- 
delocalization, dimethylammonium hydro-bis-bi-squarate, 
[(CH 3 ) 2 NH 2 «H 3 (C 4 D 4 ) 2 ], was synthesized and studied by X-ray diffraction. 
As in squaric acid, strong inter-molecular hydrogen bonding occurs. 
However, in the squarate fragment of the ammonium salt, both symmetric 
and non- symmetric hydrogen bonding are found. 

Reference: 

1. Judith K. Stalick and James A. Ibers, J. Am. Chem. Soc, 92, 

5333 (1970). — 

2. R. West and D. L. Powell, J. Am. Chem. Soc, 85, 2577 (1963). 

3. F. G. Baglin and C. B. Rose, Spect. Acta, 26 A , 2293 (1970). 



-6- 



BONDING AND STEREOCHEMICAL STUDIES OF 
GROUP IA AND IIA ORGANOMETALLIC COMPOUNDS 



Richard Zerger August 23, 1973 

In the past there has been some question as to whether or not 

hydrogen atoms are an integral part of the bonding in metal-carbon-metal 

1,2 3 4 
bridge bonds. In our attempt to answer the question the synthesis 

and structural investigation of several bicyclobutyl-organometallic com- 
pounds were undertaken. The complex bicyclobutyllithium TMEDA was found 
to be a dimer with bridging bicyclobutyl groups. This structure repre- 
sents the first example of a sclvated oligomeric alkyllithium compound 
and the first example of a tertiary carbon electron deficient compound. 
The existance of a tertiary carbon in a bridging position demonstrates 
that hydrogen involvement is not necessary for the formation of alkyl- 
metal bridge bonds. The bonding and structural properties of bis- 
bicyclobutyl-mercury as exemplified by proton nmr will also be discussed. 

The interaction of the fluorenyl anion with alkali metal cations 
has been rather extensively investigated. ' ' Brooks and co- 
workers determined the X-ray structure of f luorenyllithium bisquinu- 
clidine and interpreted their results in terms of some directed covalent 
bonding. They also predicted that in the other f luorenyl-alkali metal 
complexes containing alkali metals other than lithium, the alkali metal 
position will be determined to a greater extent by the electrostatic 
potential of the carbanion. In order to evaluate the validity of this 
prediction, the X-ray structural analysis of f luorenyl-potassium TMEDA 
was undertaken. The complex was found to be polymeric in character 
with chains of alternating anions and cations. The general aspects of 



the structure can be explained by the above electrostatic model with 
some features which may be due to either steric effects or a degree of 

directed valence bonding. 

12 
Fischer and Stolzle prepared biscyclopentadienylcalcium by the 

reaction of calcium metal with cyclopentadiene in THF . The unsolvated 
(CrHOoCa crystallized in a space group different than that reported 
for (C5H5)2Mg and (C^lL^Fe, suggesting the possibility of an unusual 
metal environment. With the hope of obtaining further information 
about the interaction between the cyclopentadienyl rings and the cal- 
cium cation, the X-ray analysis of (C5H5)2Ca was undertaken. The co- 
ordination sphere around the calcium was found to consist of two h -C5H5 
rings, one h -C5H5 ring, and one h -C5H5 ring. The first three rings 
are disposed about the calcium atom in a roughly trigonal manner with 
the fourth ring approximately perpendicaulr to the plane formed by 
the centroids of the other three rings. 

The structure of cyclohexyllithium was determined by X-ray analysis 
and found to be a hexamer. The geometry of the molecule consists of a 
cluster of six lithium atoms with the cyclohexyl rings bonded to six of 

the eight triangular faces. This structure is the same as that pre- 

13 
dieted for an alkyllithium by Brown in 1962. The bonding proposed 

for cyclohexyllithium is based on the four-centered bond theory. 



-o- 

References: 

1. K. S. Pitzer and H. S. Gutowsky, J. Amer. Chem. Soc, 68 , 

2204 (19^6). 

2. R. G. Vranka and E. L. Amma, J. Amer. Chem. Soc, 89, 3121 

(1967). 

3. S. K. Byram, J. K. Fawcett, S. C. Nyberg, and R. J. O'Brien, 

Chem. Comm., 16 (1970). 

4. F. A. Cotton, Inorg. Chem., 9, 2804 (1970). 

5. T. E. Hogen-Esch and J. Smid, J. Amer. Chem. Soc., 88, 307 

(1966). 

6. L. L. Chan and J. Smid, J. Amer. Chem. Soc, 90, 4654 (1968). 

7. T. Ellingsen and J. Smid, J. Phys. Chem., 73, 2712 (1969). 

8. R. H. Cox, J. Phys. Chem., 73, 2649 (1969). 

9. J. B. Grutzner, J. M. Lawlar, and L. M. Jackman, 

J. Amer. Chem. Soc., 94, 2306 (1972). 

10. J. A. Dixon, P. A. Gwinner, D. C. Lini, J. Amer. Chem. Soc, 

87, 1379 (1965). 

11. J. J. Brooks, W. Rhine and G. D. Stucky, J. Amer. Chem. Soc, 

94, 7339 (1972). 

12. E. 0. Fischer and G. Stolzle, Chem. Ber. , 94, 2187 (l96l). 

13. T. L. Brown, -D. W. Dickerhoof and D. A. Bafus, Amer. Chem. 

Soc, 84, 1371 (1962). 



NUCLEAR MAGNETIC RESONANCE STUDIES OF ORGANOMETALLIC 

EXCHANGE REACTIONS 

Richard L. Kieft August ?.k, 1973 

Mixtures of alkyllithium and lithium tetraalkylaluminates were 
prepared in diethyl ether and their 7 Li and proton nmr spectra ob- 
tained. Variable temperature 7 Li studies were done when R ~ methyl 
and ethyl. The slow exchange region was reached by -50°C when R = 
methyl and by -57°C when R ■ ethyl. When R - trimethylsilylmethyl 
(TMSM), 7 Li exchange is fast at -95°C The rate determining step 
for this exchange is the dissociation of the alkyllithium tetramer 
into dimers. Detailed lineshape analysis for these systems yielded 
activation energies for the dissociation of the lithium tetramer of 
12.4 kcal-mole"* 1 when R = methyl and 11.3 kcal-mole** 1 when R = ethyl. 
The activation energy when R = TMSM would be even lower indicating 
that the more electron donating alkyl groups promote the dissociation 
of the alkyllithium tetramer. 

Variable temperature proton nmr spectra were obtained for high 
ratio mixtures of MeLi:LiBr. The slow exchange region for this system 
is reached by -50°C. Detailed lineshape analysis yields a much higher 
activation energy than that obtained in the MeLi-LiAlMe 4 system, 
indicating that the rate- determining step is not the dissociation of 
the methyllithium tetramer. A new mechanism involving the dissoci- 
ation of the mixed tetramer, Li4Me 3 Br, into the dimers Li 2 Me 2 and 
Li 2 MeBr is proposed for^this system. Lineshape analysis yields a 
value of 16.5 kcal mole" 1 for the activation energy. 

Proton and 7 Li nmr studies were done on solutions containing 
various mixtures of (TMSM)Li and (TMSM)Cu. Spectra were obtained over 
the temperature interval of 44o°C to -100°C. The results indicate 
that LiCu(TMSM) 2 is the only complex formed in solution, but they do 
not reveal the extent of aggregation for LiCu(TMSM) 2 . 

Attempts were made to prepare complexes between 2-methallyl- 
lithium and bis-(trihapto-2-methallyl)nickel, but only the coupling 
product, 2,5-dimethyl-2,5-hexadiene, was obtained. The reaction of 
(trihapto-2-methallyl) nickel bromide dimer with carbon monoxide was 
examined. Infrared spectra failed to show any evidence of a di- 
carbonyl intermediate, (h 3 -C 4 H7)NiBr(C0) 2 , similar to that found in 
the analogous trihaptocyclopropenyl system. The reaction of (trihapto- 
2-methallyl) nickel bromide dimer with l,2-bis(diphenylphosphino)- 
ethane, diphos, yields the five- coordinate compound (h 3 -C 4 H7)NiBr- 
(diphos). Nmr spectra for this compound were found to be similar 
to those of the analogous palladium compound, indicating that there 
is no motion of the allyl protons leading to exchange on the nmr 
time scale up to 100°C. 



-10- 

A MOLECULAR ORBITAL DESCRIPTION OF LEWIS ACID-BASE 

INTERACTIONS 
(Final) 

Allen Marks September 4, 1973 

An equation for correlating enthalpies of adduct formation has 
recently been developed. The equation has the form: 

-AH = E aEb + C A C B (1) 

where E. and C. are empirical parameters assigned to a specific acid 
and Rq and C B are empirical parameters assigned to a specific base. 
Several correlations between thermodynamic and spectroscopic proper- 
ties have also been found recently, as well as correlations between 
E and C parameters and various molecular properties. Therefore, it 
was of interest to develop a molecular model of acid-base interactions 
with which the E and C parameters may be interpreted. 

The particular model chosen to represent the acid-base interaction 
is that of the Mulliken charge transfer model. The enthalpy of adduct 
formation is given by the energy of the initial state, before inter- 
action (minus the ionization energy of the base, I B ) minus the energy of 
the final state, after interaction (given by the energy of the lower 
energy "complex' 1 orbital). The resultant equation is 

H A + H B 2SH AB 2 V ((H A - H B )/2) 2 + P A R B 
-AH = — p^- + JZ&r- + ITs* 2I B (2) 

where It, =■ r*-H* .dT 
H AB " T* A H* B dT 
S = ^a*b^ t 

8 i = H AB " SH i 

For the particular case of reactions of neutral acids and bases 
such as 

A + B: -* A:B 

the enthalpies of adduct formation are quite small. Under these cir- 
cumstances approximations for seme of the terms in equation (2) can be 
made which simplify it to an equation similar to equation (l). It is 
found that the E.E R product represents the energy of electrostatic 
interactions between A and B whereas the C.C B product represents the 
energy gained by covalent bonding in the A:B molecule. 

For the particular base of reactions of ionic acids and bases 
such as 

M + + X" -> MX 



-li- 
the enthalpy of adduct formation is very large. Therefore, different 
approximations from the ones used above must be used to simplify 
equation (2). This treatment results in a new equation of the form: 



AH 



= \/(D A - V 2 +0 A'°B < 3) 



where D. and 0. are empirical parameters assigned to a specific acid 
and D« and R are empirical parameters assigned to a specific base. 

Equation (3) is used to correlate enthalpies of interaction of 
ionic acids and bases. Reactions of divalent acids and bases with 
monovalent and divalent bases and acids, respectively, are also in- 
cluded. A comparison is made between equations (l) and (3) in their 
ability to correlate enthalpies of reactions of both neutral and 
ionic acids and bases. 



-12- 



T1II-: REACTIVITY AND KAGNOTOCIIRMIfiTRY OP SELECTED METALLOCENES 

(Final) 

lary Ellen Switzer September 10, 1973 



The rate of ring exchange* of first transition scries metal locencs with 
liC$D 5 was studied by a near-infrared technique. Mass spectra of nickelocene 
isolated from the exchange reaction verify that the entire cyclopentadleny I 
group, rather than individual hydrogens, exchanges; the spectra Show only 
peaks corresponding to Ni(C 5 H 5 ) 2 » Ni( C 5 H 5 )( C 5 D 5 ) and NI(C s D 5 h. for M(C 5 H«) 2 
the rate of exchange decreases in the order Cr,rtn > Ni > V "» Fe,Co. The 
cobalt and iron compounds show no exchange with L » C 5D5 in four weeks or more. 
Of the first transition series metal locenes only manganocene exchanges rings 
with Ni( C5D5) 2 . The relative reactivity of the metallocenes toward ring ex- 
change parallels their reactivity toward nucleophilic substitution and can 
be explained in terms of the donor or acceptor properties of the metallocenes. 

Since the rate of exchange of nickelocene is the most favorable for 
study, the kinetics of its reaction with L t C 5D5 » the tetramethylethy lenedi" 
amine ( TMEDA) adduct of LiC 5 D 5 and Mn(C 5 Hs) 2 were investigated. A two-term 
rate law is observed for the reaction of lithium cyclopentadlenide with 
Ni( C5H5) 2. One term is first order in each reactant; the other is second 
order in LiC 5 D 5 and first order in N i( C 5 H 5 ) 2 . The rate law Is Interpreted 
as parallel reactions of monomer ic and dimeric 1 i thium cyclopentadlenide* 
In THF at 25-35° the dimerization constant is < 10"'. At 30° in THF the 
rate constants for exchange with nickelocene are 4.56xl0" 5 M* ' Sec -1 
(AH* = 11.0+3.3 kcal mole"' and AS* = -40+11 e.u.) for 1 i thium cyclopenta- 
d teniae monomers and ;> 1.5xl0" 3 M" ' sec" ' (AH* = 13.8+1.8 kcal mole" 1 and 
AS' - -28-6) for tne dimer. 



-13- 



Wo found that LiC^D 5 forms a 1:1 complex with TMEDA. The reaction of 
nickcloccnc with this complex follows a simple second-order rate law, with 
k 1.17x10"' M"' sec" 1 at 27° (AH* 1 1 . MO. 7 kcal mole" 1 and L C J -31.7.12. 1 
o. u. ) . 

Manganocene and magnesium cycl opentadi enide exchange rings slowly with 
nickeloccnc. The manganocenc exchange follows a second-order rate law, 
k - A. 35x1 0" 6 M" 1 sec"' at 30° ( AH* = 15.3^0.09 kcal mole-' and AS* = 
-30.5+2.8 e. u.). If the Ma(C 5 H 5 ) 2 exchange Is also second order, k< 
A.6xl0" 6 M" 1 sec"' at 55°. All of these ring exchange reactions are Inter- 
preted in terms of an associative mechanism in which the coordinative ability 
of the cyclopentadi cny 1 group is lower in the transition state. 

Manganocene and 1 , 1 ' -dimethyl manganocene were studied in detail. Their 
ionic properties include adduct formation with a variety of Lewis bases and 
ring exchange with nickelocene. 

The nmr data show that *.nere is some covalent character to the metal- 
ring bond in manganocene. For dimethyl manganocene the bonding is covalent 
enough for the low-spin state to become important. The existence of the 
S = 1/2 state has been deduced from the temperature-dependent magnetic 
moment of d i met hyl manganocene. At 30.5° the magnetic moment in benzene Is 
4.76 5M, well below the spin-only value of 5 92. At: 4.2°K esr experiments 
snow that the ground stave Is 2 £ (gii = 2.922, gi = 1.893 in methylcyclo- 
hexane glasses). The existence of a *A state a: 138°K in a toluene glass 
is inferred from the isotropic g value at this temperature. The low spin 
2 A state is used *z a tentative explanation of the large downfield shifts 
in the nmr of dimethyl Manganocene at 310°K (-137 and -77 ppm in benzene for 
l he ring and methyl rt^onances, respectively), 



-14- 



SPECTROSCOPIC STUDIES OF PENTA- COORDINATE 
ORGANOMETALLIC COMPLEXES 

Cheryl D. Pribula September 13, 1973 

Isoelectronic penta-coordinate complexes [M(C0) 5 Y P v ] n , where 

P = P(C Q H 5 )3 or P(OCH 3 ) 3 , and M = Mn (n = -l), Fe = (n = 0), or 
Co (n = +l), have been studied using nqr, Mbssbauer, ir and nmr 
spectroscopy. 

Small changes in metal valence electron distribution are re- 
flected by the field gradient at the metal nucleus, ec U z » This 
parameter was determined from nqr ( 55 Mn, 59 Co) and MosSDauer ( 5T Fe) 
experiments for a number of complexes. 

To a good approximation, ec L 77 is given by 

eq zz = eq 32 orN d + 1/2 (H + N d ) - (H fl + N )] 

z 2 xz yz xy x^-y 

+ eq 410 rN - 1/2(H + N )] (l) 

1 z p x p y 

The N' s are orbital populations, usually approximated by Mulliken 
population analysis methods. Values of eq 320 and eq 410 > which are the 
field gradients due to a single electron in a 3d and 4 orbital, 

z 2 p z 
respectively, were calculated for the metal ions Mn(-l), Fe(o), and 
Co(+l) in different electron configurations. Although it has usually 
been assumed that the p-orbital contribution is insignificant, we have 
shown that this term is comparable in magnitude with the d term. For 
Fe(C0)5, eq was evaluated from equation (l). The results are in 
good agreement with the experimental field gradient. Relative values 
of eq 32 o for the three metal ions are compared with relative values of 
eq observed for [MnfCOJs]" and Fe(CO)s and estimated for the hypo- 
thetical [Co(CO) 5 ] . It is concluded that the populations of the 
planar d orbitals increase, relative to the axial, in the order 
Mn < Fe < Co. 

The effect of phosphorus ligand substitution on eq was examined. 
For positive eq (known for the Fe complexes and assumed for Mn and 
Co), substitution of the stronger a-donating poorer n-accepting 
(relative to CO) P ligand should lower eq by increasing the popu- 
lations of d and d„ . Actually a simple trend is not observed, 
z 2 xz,yz 

and the results are not easily interpretable. 

Infrared spectra of LiMn(CO) 5 and NaMn(CO) 5 in the CO stretching 
region have been examined in ether solvents. The spectra are attributed 
to an ion-paired species of C 3 symmetry in which alkali metal is bonded 
to carbonyl oxygen. Treatment of NaMn(CO) 5 /THF with Mg 2 gives a 
species whose ir is consistent with a Mg-OC interaction. 



-15- 



The complex [Co(C0) 2 (P(0CH 3 )3] was examined from -105° to +40° 
using several different nmr techniques. The room temperature pmr 
spectrum indicates rapid intramolecular exchange, while at -105°* 
exchange is apparently slow. The complexity of the spin system pre- 
cludes mechanistic analysis. Other techniques, 31 P-decoupled 1 H 
nmr and -"-H-decoupled 31 P nmr, produce fast exchange spectra even at 
low temperatures. At 30°, the 13 C nmr exhibits a very broad carbonyl 
resonance due to 59 Co- 13 C interaction, although at lower temperatures 
this effect is negligible. A method of decoupling 59 Co must be 
devised if 13 C is to be useful in mechanistic studies. 



KINETICS AND MECHANISMS OF THE SUBSTITUTION REACTION OF ALKYL- 
GROIJP IV METAL COBALT -TETRACARBONYLS 

James Eaton September 18, 1973 

The reaction of Ph 3 MCo(CO) 4 (M = Si, Ge, !)n) was studied in the 
presence of phosphine bases. The rate of reactivity was shown to de- 
pend on both the incoming base and the acidity of the group IV metal. 
The rate of reactivity was Si » Ge > Sn. The reaction was shown to 
obey the rate law: 

rate = k 2 [L] FPh 3 MCo(CO) 4 1 

where L is the Lewis base concentration. 

Several correlations were made between the acidity of the group 
IV metal and the rate of reaction. Based on this, the initial step in 
the mechanism was proposed to be base attack at the group IV metal. 

The silicon reaction was studied quite extensibly in order to 
determine the complete mechanism. When the base was P(C 4 Hc,)3j the 
initial product was [Ph 3 S2?(C 4 H9) 3 ] [Co(C0) 4 ]~. This rearranged upon 
standing to give the final product trans-Ph 3 SiCo(CO) 3 P(C 4 H<)) 3 . This 
reaction was not affected by the presence of excess CO. On the other 
hand, when the base was PPh 3 , no ion pair was detected. The product 
again was trans-Ph 3 ";*Co(C0) 3 PPh3. The presence of excess CO caused the 
rate of this reaction to slow down. Based on these observations, a 
mechanism was proposed. It involved base attack at the group IV metal 
followed by ion pair formation. This ion pair then recombined to give 
a reactive intermediate in which cobalt was coordinatively unsaturated. 
This intermediate then rearranged to yield the final product. 

The rate constants measured involve only the base attack at the 
group IV metal. This is, however, the rate determining step. 



-17- 



ELECTRON TRANSFER REACTIONS OF COBALT COMPLEXES 



Michael Broccardo September 18, 1973 

The decrease in the specific rate of outer-sphere electron trans- 
fer reactions observed when ammonia ligands in the inner-coordination 
sphere of a transition metal ion are replaced by ethyl enedi amine has gen- 
erally been explained as due to the greater size and steric bulk of the 
latter ligand. However, hardly any systematic effort "has been directed 
toward studying complexes containing saturated ligands bulkier than 
ethyl enedi amine with the purpose of determining if such ligands are able 
to further reduce the specific rate of electron transfer, i.e. to serve 

as "electron insulators". Therefore, in this study the outer-sphere 

3+ 
electron transfer reactions of Co(chxn) 3 (chxn = trans-1 ,2-diamino- 

3+ 
cyclohexane) and Co(en) 3 (en = ethyl enedi amine) with a common reductant 

have been examined in detail. 

The observed rate constants at 25 C and associated activation pa- 

3+ 3+ 

rameters for the reactions of Co(chxn) 3 and Co(en) 3 with 

Ru(NH 3 ) 5 H 2 2+ in aqueous 0.10 M CI" media are k = 3.8 X 10" 3 M" 1 sec' 1 , 

AH* = 17.6±0.5 Kcal/mole, AS* = -10.6±le.u. and k = 6.2 X 10" 3 M" 1 sec" 1 , 

AH = 17.1±0.5 Kcal/mole, AS* = -11.3±le.u., respectively. The actual 

reductant in this system is shown to be Ru(NH 3 ) 5 Cl . 

2+ 
For V as the reductant in aqueous 0.10 M CF 3 C0 2 media, the spe- 

3+ 
dfic rate constant and associated activation parameters for Co(chxn) 3 

are k = 5.8 X 10" 4 M' 1 sec" 1 , AH* = 14.8±0.9 Kcal/mole, and AS* = 



-18- 



-25.412.9 e.u.; and for Co(en) 3 3+ , k = 9.6 X 10" 4 M" 1 sec" 1 , AH* = 
14.2±0.3 Kcal/mole, and AS* = -26.4±0.9 e.u. 

It is concluded that the mere size or bulk of the ligand coordi- 
nated to the cobalt(III) metal center has only a small effect upon the 
rate of outer-sphere electron transfer. In addition, the lack of any 

significant difference in AS for the two oxidants suggests that the 

3+ 
reactions of Co(chxn)~ are not dramatically less adiabatic than those 

3+ 
of Co(en) 3 . 

The second part of this report examines the rate of outer-sphere 
electron exchange between the cobalt clathro chelates, [Co (dmg) 3 (BF) 2 ] 
and [Co (dmg) 3 (BF)«], in which the central cobalt ion is coordinated to 
three dimethyl gloxime (= dmg) ligands through six nitrogen atoms. Numer- 
ous earlier investigations have demonstrated that the rates of outer- 
sphere electron transfer between cobalt(III) and cobalt(II) complexes 
are quite slow in comparison with the rates of self -exchange between 
analogous complexes of other transition metals. 

Since octahedral cobalt(II) complexes are usually high-spin with a 

4 
T ground state while cobalt(III) complexes are almost always low-spin 

1 
with a A ground state, this anamolous behavior has frequently been 

attributed to the changes in spin multiplicity which occur with electron 

transfer. As both of the cobalt clathro chelates are low-spin species, 

the self-exchange electron transfer reaction is "spin-allowed", i.e. 

electron exchange can take place without a spin multiplicity change on 

either reactant in the activated complex except for the transferred 

electron. 



-19- 



Al though the exact value for the rate of self-exchange was not 
determined, it was demonstrated via exchange broadening of the nmr spec- 
trum of the diamagnetic cobalt(III) complex that k ,.< 400 M~ sec" at 
80° in nitromethane. This is not appreciably faster than the rates of 
electron transfer observed for cobalt(III)-cobalt(II) systems which are 
"spin-forbidden", i.e. which involve a spin-multiplicity change. It is 
concluded that outer-sphere cobalt(III)-cobalt(II) self-exchange re- 
actions are inherently slow because of the considerable coordination- 
sphere and solvation-sphere reorganization energies which accompany 
transfer of an anti -bonding electron and not because of spin multiplicity 
changes. 



OXIDATIVE DEHYDROGENATION AND KINETICS OF 
RUTHENIUM (II ) -ETHYLENEDIAMINE COMPLEXES 

Dennis Mahoney October 5, 1973 

Oxidative dehydrogenation of large macrocyclic ligands has 
been known to occur for a number of years. More recently it has 
been found that small bidentate ligands such as ethylenediamine 
and even monodentate ligands such as methylamine may also react 
in this fashion. In all cases the ligand remains coordinated 
throughout the process: 

Ox + M(II)-en ):i(II)-diim + 4H + 

Here M(II) represents a divalent metal whose valence state re- 
mains unchanged and en and diim refer to ethylenediamine and its 
diimine analog. 

In the studies reported here the oxidation of ruthenium (II ) - 
ethylenediamine complexes was studied as a function of the acid 
strength and the reduction potential of the complex. The com- 
plexes studied included Ru(en) 2+ (E°=+0.19V), Ru (en) 2 phen 2+ (E°= 
+0.55V), and Ru (phen) 2 en 2+ (E o =+0 . 76V) . Oxidation will yield 
either the ruthenium (III) -ethylenediamine complex or the rutheni- 
um (II) -diimine complex depending on the acid strength. As the 
reduction potential of the complex becomes (numerically) larger, 
stronger acid strengths are required to prevent oxidative dehydro- 
genation. Participation of the metal in the oxidative dehydrogena- 
tion process is also implied by the cyclic voxtammetry of the com- 
plexes. As the E° increases the potential required to induce oxida- 
tive dehydrogenation increases but is always higher than that ne- 



-2- 

cessary to reversibly oxidize the metal. These results are in ac- 
cord with those found in macrocyclic compounds wherein the metal 
is first oxidized and is then reduced in attacking the ligand. 

In the second part of the study the electron transfer rate 

2+ 3+ 

of Ru(en)pdiim reacting with Fe was investigated in an effort 

to determine the effects of ligand conjugation on electron trans- 
fer rate . The rate is found to be much slower, than that of the 

2+ 
saturated analog, Ru(en), , under the same conditions. The re- 
sults are discussed in terms of the Marcus Theory of Electron 
Transfer. The difference is shown to be due to a substantially 
larger free energy change in the latter case. It is concluded 
that the effect of conjugation is small, at least in these sys- 
tems, due to the high transmission coefficients present, even in 
the saturated systems. 



-22- 

BIOCHEMICAL APPLICATIONS OF LANTHANIDE SHIFT REAGENTS 

Pamela A. Milton November 20, 1973 

INTRODUCTION 

The application of lanthanide shift reagents (LSR's) to nmr 
spectroscopy has proven to be an important outgrowth of the more general 
usage of paramagnetic shift reagents. 1 In the few years since Hinckley 2 
first systematically applied the LSR's to nmr spectral clarification, 
the field has rapidly expanded and now includes such applications as 
determination of optical purity, elucidation of conformation and struc- 
ture of compounds in solution, dynamic nmr spectroscopy and the most 
frequently employed simplification of complex, overlapping spectra. 3-7 
LSR's have been used most extensively in resolving spectra of organic 
molecules and more recently have been applied to organometallic mole- 
cules and biological model systems. 

THEORY 8 " 12 > 12a 

The LSR interacts with the compound to be studied by coordinating 
with donor atoms of the molecule, thus forming a new complex in so- 
lution. This coordination effects a "shift" in the nmr spectrum of the 
compound of interest by inducing a large local magnetic field due to the 
unpaired electronic spin of the paramagnetic lanthanide ion. Variations 
in the local magnetic field introduce additional magnetic non-equiva- 
lences of the surrounding nuclei. 

The shift consists of a contact and a pseudocontact contribution. 
In the case of the lanthanide ions, the pseudocontact contribution 
predominates. 

Structural and conformational studies in solution have made use of 
the predominance of the pseudocontact interaction in the lanthanide ions 
in that the pseudocontact mechanism depends directly on the magnetic 
anisotropy of the complex formed and on geometric factors between the 
metal ion and the coordinated nuclei. 

Added structural information may be gained through the use of 
certain LSR's as broadening or relaxation reagents. Lanthanides with 
this broadening ability possess a long electronic relaxation time as 
compared to the molecular reorientation times. 

BIOCHEMICAL APPLICATIONS 13 ' 22 ~ 24 

Since extensive application has been made of LSR' s to studies of 
conformation and structure of organic compounds in solution, a natural 
outgrowth of this area has been consideration of biologically important 
molecules. Much recent work has been done on conformational studies 
of mono- and dinucleotides in solution. 14 " 16 The intent of such studies 
is to derive information on the effect of environmental conditions on 
the conformation of molecules in order to be able to predict and ration- 
alize the conformation of molecules in biological systems once the 
conditions of the local biological environment are known. 



-23- 

One of the more significant applications of LSR's over the last 
few years has been their use as probes of membrance structure. 1T ~ 21 
Most of this work has been done on model membrane systems consisting 
of sonicated phospholipid bilayers. NMR techniques using paramagnetic 
ion probes have been able to differentiate between the internal and 
external phospholipid monolayers. 

In biological membranes the added presence of lipoproteins very 
likely produces an asymmetric distribution of the various phospholipid 
classes in the two surfaces of the membrane. The ready associability 
of solvated paramagnetic ions with lipid polar headgroups in lipid 
vesicles provides a potential method for determining the distribution 
of lipid classes in the surface of biological membranes. 

Enzymatic activity may be very dependent on the binding of the 
protein and/or the substrate in the lipid bilayer. There may also be 
a correlation between the motional factor in the lipid bilayer and en- 
zymatic activity. Since both relaxation and shift reagents supply 
information on distance and orientation relative to the phosphate head 
groups and the membrane surface, lanthanide ions should prove to be 
particularly well-suited to such studies as the above. 

Other contexts in which the LSR's should serve to be applicable 
include studies of electron transport intermediates, the structure and 
conformation of metalloproteins and the mechanism of transport of 
various molecules through cell walls. 

REFERENCES 

Reviews 

1. R. von Ammon, R. D. Fischer, Angew. Chem. Internat. Edit., 11 , 
675 (197?). 

2. C. C. Hinckley, J. Amer. Chem. Soc, 91, 5l6o (1969). 

3. Nuclear Magnetic Resonance Shift Reagents , R. E. Sievers, ed. , 
Academic Press, New York ( 1973) . 

4. J. Reuben, Prog. NMR Spec., 9, 1 (1973). 

5. B. C. Mayo, Chem. Soc. Rev., 2, 49 (1973). 

6. J. K. M. Sanders, D. H. Williams, Nature, ?40, 385 (1972). 

7. M. R. Peterson, Jr., G. H. Wahl, Jr., J. CJHem. Ed., 49, 790 (197?). 

Theory 

8. H. M. McConnell, R. E. Robertson, J. Chem. Phys., 29, 136l (1958). 

9. R. J. Kurland, B. R. McGarvey, J. Magn. Reson. , 2, 286 (1970). 

10. B. Bleaney, J. Magn. Reson., 8, 91 (1972). 

11. B. Bleaney, C. M. Dobson, B. A. Levine, R. B. Martin, 

R. J. P. Williams, A. V. Xavier, Chem. Comm. , 791 (1972). 

12. A. Abragam, B. Bleaney, Electron Paramagnetic Resonance of 
Transition Ions , Clarendon Press, Oxford (1970), Chapt. "57 

l?a. W. DeW. Horrocks, Jr., J. P. Sipe, III, D. Sudnick, p. 53, 

Nuclear Magnetic Resonance Shift Reagents , R. E. Sievers, ed. , 
Academic Press, New York (1973) . 



-24- 



General Biochemical Applications 

13. J. A. Glasel, Prog. Inorg. Chem. , 18, 383 (1973). 

14. C. D. Barry, A. C. T. North, J. A. Glasel, R. J. P. Williams, 
A. V. Xavier, Nature, 232, 236 (1971). 

15. C D. Barry, J. A. Glasel, A. C. T. N rth, R. J. P. Williams, 
A. V. Xavier, Biochim. Biophys. Res. Comm. , 47, 166 (1972). 

16. C. D. Barry, J. A. Glasel, A. C. T. North, R. J. P. Williams, 
A. V. Xavier, Biochim. Biophys. Acta, ?£>o , 101 (1972). 

Membranes 



17. 

IB. 
19. 

20. 
21. 



V. F. Bystrov, N. I. Dubrovina, L. I. Barsukov, L. D. Bergelson, 
Chem. Phys. Lipids, 6, 343 (1971). 
R. J. Kostelnik, S. M. Cactellano, J. Magn. 
S. B. Andrews, J. W. Falle r, J. M. Gilliam, 
Proc. Nat. Acad. Sci. USA, 70, l8l4 (1973). 
Y. K. Levine, A. G. Lee, J. M. Birdsall, J. 
Robinson, Biochim. Biophys. Acta, 291 , 

C. 



Reson. , 7, 219 (1972) 
R. J. Barnett, 



J. D. 

Membrane Molecular Biology , 



Sinauer Assocs. , Stamford (1972). 



F. Fox, A. 



C. Metcalfe, 
592 (1973). 
Keith, eds., 



Lysozyme Studies 



22. 


K. 


G. 




A. 


V. 


23. 


R. 


A. 




R. 


J. 


24. 


C. 


G. 




R. 


J. 



Morallee, E. Nieboer, F. J. C. Rossotti, R. J. P. Williams, 

Xavier, Chem. Comm., 1132 (1970). 

Dwek, R. E. Richards, K. G. Moral] ee, E. Nieboer, 

P. Williams, A. V. Xavier, Eur. J. Biochem. , 21, 204 (1971) 

Butchard, R. A. Dwek, S. J. Ferguson, P. W. Kent, 

P. Williams, A. V. Xavier, Febs Lett., 25, 91 (1972). 



-25- 



NUCLEAR RELAXATION PROCESSES: APPLICATIONS 
Dennis R. Kidd November 27, 1973 

Although the theory for nuclear spin relaxation has been known 
since 19^8, 1 measurement of relaxation times was difficult until the 
advent of Fourier transform nmr instruments. The three relaxation 
times, Ti, T 2 , T 2 *, are measured as follows: 

l/T 2 * = tt v x / 2 

(v 1( / 2 is the line width at half maximum intensity of a normal ab- 
sorption spectrum. ) 

Ti by a pulsed experiment using a (l8o°-T-90°) sequence or a 
(T-90°-T-180°-t-90°) sequence. 

T 2 by using a (90°-t-18o°) sequence. 

There are five types of interaction which contribute to Ti and 
T 2 : 2 > 10 

1. magnetic dipole-dipole 

2. spin rotation 

3. chemical shift anisotropy 

4. electronic quadrupole 

5. scalar coupling of the first and second kind 

Each interaction gives rise to fluctuating magnetic fields which 
couple the spin system to the lattice. This is expressed by the matrix 
element 

(s f|AF(t)|s' f ' ) = (s |A|s' )(f |F(t)| f ) 

(where f and f are lattice functions, s and s' are spin functions, 
A is a spin operator and F(t) is a lattice operator. A and F depend 
on the interaction involved. ) 

Using the appropriate matrix element and the standard method for 
calculation of transition rates, one obtains the expression for the 
rate of relaxation by each interaction. These expressions are given 
below for when the extreme narrowing condition is met. 

magnetic dipole-dipole 3 

(iAx) rot = (iA 2 ) rot = g ^ I+1 > t c 



-?<>- 



chemical shift anisotropy 4 

2 

15 



1/Ti = yf Y 2 Ho P (ct 11 -o 1 ) 2 t ( 



1/T 2 = 7^ Y 2 H 2 (a 11 -a J _) 2 T c 
spin rotation 

i/ Tl = Prr I^£ C 2 

x rp erf j 

electric quadrupole 

1/Tl . l^«- lf §^ r 2 rT (l + ^) (5jfc)«T c 

scalar coupling of the first kind 

p t 

1/iPi = ■* A 2 S (S + 1) h— - -, - ^ — ?} 

_L o e 

1/T 2 = §h- S (S + 1) [t + T e 1 

1 + (uj t -(i)„ j^re^ 

scalar coupling of the second kind replace t with t . 

Applications 

The applications of nuclear relaxation times have recently 
been reviewed. 5 

J. Jagur-Grodzinski, et al. , studied the complexation reaction 
of sodium ions with crown ethers in various solvents using 23 Na nmr 
spectroscopy. There is almost no change in the chemical shift of 23 Na 
upon complexation by the crown ether. However, complexation could be 
observed by noting the difference in T 2 due to quadrupole relaxation. 6 

In 1966, T. R. Stengle and J. D. Baldeschwieler proposed the use 
of mercuric chloride to probe the sulfhydryl sites in proteins and to 
determine the correlation time of the bound halide. Later, E. J. Wells, 
et al . , showed that using mercuric bromide as the probe permitted one 
to gain kinetic data relating to the halide mercury exchange. Y,B 

13 C Ti measurements have been used to determine a specific elec- 
tron-nuclear relaxation rate which is used for investigating solvent- 
solute interaction with paramagnetic Cr(acac) 3 and to follow segmental 
motion in carbon chains. 



- 



REFERENCE 

N. Bloembergen, E. M. Purcell, anci R. V. Pound, Phys. Rev., 73, 

679 (19 ] ! 0. ~ 

a. T. C. Farrar and E. D. Becker, "Pulse and Fourier Transform 
NMR," Chapter 4, Academic Press, Inc., New York, 1971; 

b. A. Abragam, "The Principles of Nuclear Magnetism," Chapter 8, 
Oxford University Press, London and New York, 196] ; 

c. A. G. Redfield, Adan. Magn. Resonance, 1, 1 (19 ( ">5). 

J. H. Noggle and R. E. Schirmer, "The Nuclear Overhauser Effect," 
Chapter 1, Academic Press, Inc., New York, 1971. 

a. G. C. Levy, D. M. White, and F. A. L. Anet, J. Magn. 
Resonance, 6, 453 (197?); b. T. C. Farrar, S. J. Druck, 
R. R. Shoup, and E. D. Becker, J. Am. Chem. Soc, 94, 699 (197?). 

a. G. C. Levy, Accounts Chem. Res., 6, l6l (1973); 

b. T. C. Farrar, A. A. Margott, and M. S. Malmberg, 
Ann. Rev. Phys. Chem., 23, 193 (197?). 

a. E. Shchori, J. Jagur-Grodzinski, Z. Luz, and M. Shporer, 
J. Am. Chem. Soc, 93, 7133 (1971 ); 

b. E. Shchori, J. Jagur-Grodzinski, and M. Shporer, 
J. Am. Chem. Soc, _95, 384? (1973). 

T. R. Stengle and J. D. Baldeschwieler, Proc. Nat. Acad. Sci. 
U. S., 55, 10?0 (1966). 

T. R. Collins, Z. Starcuk, A. H. Burr, and E. J. Wells, 
J. Am. Chem. Soc, 95, l649 (1973). 

a. G. C. Levy and J. D. Cargioli, J. Mag. Resonance, 10, ?31 
(1973); b. G. C. Levy, J. D. Cargiolo, and F. A. L. Anet, 
J. Am. Chem. Soc, 95, 1527 (1973). 

C. P. Poole, Jr., and H. A. Farach, "Relaxation in Magnetic 
Resonance," Academic Press, Inc., New York, 1971. 



- S\)- 



MANGANESE PORPHYRINS AND PHTHALOCYANINES 
Lynne M. Parr February 5, 197^ 

Metals are essential elements in a number of biological pro- 
cesses. Their role ranges from trigger and control mechanisms to 
structural contexts and redox catalysis. 1 Manganese is present in 
many of these roles. Many enzymes require manganese, and a few 
manganese- containing proteins are known. 2 

One of the important roles of manganese is in the oxygen evo- 
lutionary process of photosynthesis. 3 The nature of the manganese 
complex in the chloroplast is unknown, however, it may be a porphyrin 
or chlorophyll- like molecule. Model studies have thus been under- 
taken on these types of systems in an effort to understand the role 
of manganese in photosynthesis. 

MANGANESE PORPHYRINS 4 ' 5 ' 6 

Manganese porphyrins are most stable as Mn(lll). These complexes 
contain a complexed anion, exhibit no ESR signal, and have a magnetic 
moment indicative of a high-spin d 4 complex. The electronic spectrum 
of the Mn(lll) porphyrin complexes is unusual. No Soret band is ob- 
served. Rather, the spectrum consists of six bands whose positions 
vary with the porphyrin ligand used, the solvent, and the anion. 

The Mn( III) -porphyrins may be reduced to the corresponding 
Mn( II ) -porphyrin, or oxidized to the Mn( IV") -porphyrin. The redox po- 
tentials vary markedly with pH and porphyrin. Some of the reduction 
potentials are large enough to cause the oxidation of water. 

The manganese porphyrins are similar to the iron porphyrins in 
many ways. They also exhibit some unique differences. Many of the 
differences arise from the fact that Mn(lll) is only found in the 
high- spin state. 

MANGANESE PHTHALOCYANINES 7 ' 8 ' 9 

These complexes are similar to the porphyrins in many ways. 
They do, however, exhibit some interesting differences. The chemistry 
of these complexes is quite extensive. An oxo-bridged species has 
been investigated and has sparked considerable controversy. 

The observed chemical behavior of these groups of compounds has 
led to a proposed mechanism for the evolution of oxygen from plants 
during photosynthesis^ 



-P9- 



References 

1. M. N. Hughes, "The Inorganic Chemistry of Biological Processes, 

Wiley and Sons, New York (197?). 

2. D. C. Borg, G. C. Cotzias, Nature, 18?, 1678 (58). 

3. Cheniae and Martin, Plant Physiol., 44, 351 (1969). 

4. Melvin Calvin, Rev. Pure and Appl. Chem., 15, 1 (1965). 

5. L. J. Boucher, Coord. Chem. Rev., 7, 289 (1975?). 

6. Loach and Calvin, Biochem. , 2, 36l (1963). 

7. G. Engelsma, et al., J. Phys. Chem., 66, 2517 (196?). 

8. A. Yamamoto, et al., Inorg. Chem., 7$ 847 (1968), 

9. G. W. Canham and A. B. P. Lever, Inorg. Nucl. Chem. Lett., 9> 

513 (1973). 

General 

10. Levason and McAuliffe, Coord. Chem. Rev., 7, 353 (197?). 

11. J. H. Wang, Ace. Chem. Res., 3, 90 (1970). 
1?. J. M. Olson, Science, 168, 438 (1970). 



-30- 



THE ACTIVE SITE OF MEMERYTHRIN 

Mamoru Tachikawa February 12, 197 ^ 

Introduction 

Hemerythrin is an oxygen carrier pigment ,found in some in- 
vertebrate phyla: sipunculids, polychaetes, priapulids, and bachiopods. 
This pigment is known to exist as an octamer. Each subunit or monomer 
which contains two non-heme iron atoms reversibly combines with one 
oxygen molecule. 1 To elucidate the structure of the active site of 
hemerythrin, particularly the oxidation states of the two iron atoms 
and their relative positions as well as that of ligands coordinating 
to the metal, magnetic susceptibility measurements, U.V., visible, CD., 
O.R.D., 2 ' 3 ' 4 Mttssbauer, and resonance Raman spectroscopies, and amino 
acid sequencial, compositional and reactivity analyses have been 
employed. 5 

Deoxyhemerythrin 

In deoxyhemerythrin the iron atoms seem to exist as magnetically 
independent high- spin iron(ll) according to MBssbauer and magnetic 
susceptibility measurements. 6 ' 7 The two iron atoms can either combine 
with an oxygen molecule forming oxyhemerythrin, or be oxidized by an 
oxidizing agent such as potassium hexacyanoferrate(lll) and then com- 
bine with one or two ligands forming methemerythrin. 8 

Oxyhemerythrin 

Oxyhemerythrin could be considered as either two iron(ll) atoms 
combined with an oxygen molecule or two iron(lll) atoms combined with 
a peroxy ion. The spectroscopic and magnetic measurements support the 
latter, where two high- spin iron(lll) are strongly antiferromagnetically 
coupled. 6 * 7 ' 9 * 10 ' 11 Iron Mossbauer spectra of oxyhemerythrin show two 
doublets indicating that there are two non- equivalent iron atoms in a 
subunit, while only one type of iron is observed in Mossbauer spectra 
for deoxy- and methemerythrin. 12 

Methemerythrin 

Methemerythrins have been subject to active investigations, 
though they are not naturally occurring active forms of this protein. 
In this class of compounds, two high- spin iron (ill) atoms are also 
strongly antiferromagnetically coupled. 10 ' 11 Anions such as azide, 
thiocyanate, cyanide, isocyanate, bromide, chloride, fluoride, and 
hydrosulfide are known to form complexes with an iron to anion ratio 
of 2 to 1, in which the two iron(lll) atoms are indistinguishable in 
iron Mftssbauer spectra. 7 ' 12 This implies that the anions may be bridg- 
ing the two iron atoms in a symmetrical manner. Water, hydroxide, and 
fluoride ions coordinate to iron in the ratio 1 to l. 13 



-31- 



Coordinating Groups 

Since hemerythrin does not contain any heme group, the co- 
ordination sphere of the iron atoms has to be completed by amino acid 
side chains, water or other small molecules or ions. Chemical modi- 
fication of amino acid side chains by specific reagents indicates that 
there are four slow- reacting histidine and two slow- reacting tyrosine 
residues. 14,15,16,17 Tft e only cysteine residue found in a subunit is 
not coordinated to iron. 18 The terminal as well as e-amino group of 
the lysine residues are also believed not to be the coordinating 
group. 14 Comparison of amino acid sequences of hemerythrin from dif- 
ferent species could be a powerful method to find out the essential 
residues for the active site. Sequences of two hemerythrins, that of 
Dendrostomum pyroides and Golfingia gouldii are known, but they are 
too close to each other for this purpose. 19 * 20 * S1 * 22 

Model Systems 

From magnetic and Mossbauer evidence, fi-oxo high- spin di- 
iron(lll) compounds have been proposed as a model system of the active 
site of the oxygen carrier protein. Based on this model, several 
spectroscopic features of hemerythrin have been predicted, but they 
have not been unambiguously proved due to hindrance by the protein 
absorption bands. 23 Several structural models for oxyhemerythrin have 
been suggested to explain the non- equivalence of the two iron sites. 
However, there is no definitive evidence for any of the models. 11 * 13 

References 

1. Klotz, I. M. , Klotz, T. A., Fiess, H. A., Arch. Biochem. Biophys., 

68, 284 (1957). 

2. Darnall, D. W. , Garbett, K. , Klotz, I. M. , Arch. Biochem. 

Biophys., 133, 103 (1969). 

3. Bossa, F. , Brunori, M. , Bates, G. W. , Antonini, E., Fasella, P., 

Biochim. Biophys. Acta, 207, 41 (1970). 

4. Klippenstein, G. L. , Van Piper, D. A., Oosterom, E. A., 

J. Biol. Chem., 247 , 5959 (1972). 

5. Okamura, M. Y. , and Klotz, I. M. , in Eichhorn, G. L. , 

Inorganic Biochemistry, Vol. 1, Elsevier, New York, 1973, 
pp. 320-343. 

6. Okamura, M. Y. , Klotz, I. M. , Johnson, C. E. , Winter, M. R. C. , 

Williams, R. J. P., Biochemistry, 8, 1951 (19&9). 

7. York, J. L., Bearden, A. J., Biochemistry, 9, 4549 (1970). 



-32- 



8. Keresztes-Nagy, Klotz, I. M. , Biochemistry, ^4, 919 (1965). 

9. Dunn, J. B. R. , Shriver, D. F. , Klotz, I. M. , Proc. Nat. 

Acad. Sci. USA, 70, 2582 (1973). 

10. Moss, T. H. . Moleski, C. , York, J. L. , Biochemistry, IX), 

840 (1971). 

11. Gray, H. B. , Dawson, J. W. , Hoenig, H. E. , Rossman, G. R. . 

Schredder, J. M. , Wang, R. , Biochemistry, JUL, 46l (1972). 

12. Garbett, K. , Johnson, C. E. , Klotz, I. M. , Okamura, M. Y. , 

Williams, R. J. P., Arch. Biochem. Biophys., 142, 574 (1971). 

13. Garbett, K. , Darnell, D. W. , Klotz, I. M. • Williams, R. J., 

Arch. Biochem. Biophys., _135, 419 (1969). 

14. Fan, C. C. , York, J. L. , Biochem. Biophys. Res. Commun. , 36 , 

365 (1969). 

15. Rill, R. L. . Klotz, I. M. , Arch. Biochem. Biophys., 147 , 

226 (1971). 

16. York, J. L., Fan, C. C., Biochemistry, 10, 1659 (1971). 

17. Fan, C. C., York, J. L. , Biochem. Biophys. Res. Commun., 47, 

472 (1972). 

18. Keresztes-Nagy, Klotz, I. M. , Biochemistry, 2, 923 (1963). 

19. Klippenstein, G. L. , Holleman. J. W. , Klotz, I. M. , 

Biochemistry, 7, 3868 (1968). 

20. Ferrell, R. E. , Kitto, G. B. , Biochemistry, 10, 2923 (1971). 

21. Klippenstein, G. L., Biochemistry, 11, 372 (1972). 

22. Joshi, J. G. , Sullivan, B., Comp. Biochem. Physiol., 144b , 

857 (1973). 

23. Gray, H. B. , Advances in Chemistry Series 100, pp. 365-389 

(1971). 



-33- 

HOMOGENEOUS INORGANIC OSCILLATING REACTIONS 
Ben Fieselmann March 5, 197^ 

Periodic fluctuations in the measurements of quantities such as 
concentrations, absorbance, electric potential, and the rate of gas 
evolution have been reported in a significant number of chemical 
reactions. 1 Glycolysis in yeast cells, 2 hydrocarbon combustion, 3 
dissolution of chromium in hydrochloric acid, and numerous other widely 
varying types of reactions produce oscillations. Only within the last 
few years have studies begun to explain the causes of oscillations and 
the mechanisms. 4 The relative ease of collecting data and the availa- 
bility of relevant kinetic studies make homogeneous inorganic oscil- 
lating reactions the most favorable systems for studying the general 
phenomena of oscillation in chemical systems. 

Thermodynamic Considerations 

The thermodynamics of oscillating systems have well developed by 
physicists and mathematicians. 1 Oscillating reactions proceed with a 
net reaction and a decrease in Gibbs free energy. 5 Unlike reactions 
which monotonically approach equilibrium, oscillating systems fluctuate 
around quasi-steady state equilibria as they advance slowly toward true 
equilibrium. As long as a reaction moves toward true equilibrium, there 
is no restraint that prevents oscillation in the concentrations of 
intermediates; however, oscillation about true equilibrium is theo- 
retically forbidden. 

Kinetics of Oscillation 

The kinetics of oscillating systems can be described as the coup- 
ling together of several separable reactions. One prerequisite for 
oscillation in a chemical system is an autocatalytic step that functions 
as a feedback switch to turn on and off the reaction step giving rise 
to the observed fluctuations. 7 The reactions must be far from equilib- 
rium for oscillations and must approximate an open system in which re- 
actants effectively come from a large reservoir and products are deposi- 
ted into a sink. 8 Rate constants for individual steps must fall within 
certain limits to allow the required coupling of reactions. 

The theoretical possibility of undamped chemical oscillations was 
first proposed by Lotka in 1920. 9 With the advent of computers, rate 
expressions derived from numerous theoretical mechanisms have been 
shown to predict periodic fluctuations in concentrations. 10 

Belousov Reaction 

The best understood homogeneous oscillating reaction is the 
Belousov reaction. 11 When a bromate salt is mixed with the required 
concentrations of cerium(lll) and malonic acid in an aqueous sulfuric 
acid solution, oscillations occur in bromide and cerium(lll) concen- 
trations. Temperature gradients and spatial oscillations have been 



-34- 

reported. 12 The net reaction is the following: 13 

3H + + 3Br0 3 ~ + 5CH 2 (C00H) 2 ■+ 3BrCH(C00H) 2 + 2HC00H + ^C0 2 + 5H 2 

Noyes and coworkers proposed a ten step mechanism for the Belousov 
reaction that is consistent with observed periodic changes in concen- 
trations, with available redox potentials, and with kinetic studies in 
other laboratories. 11 Their method was to break the full reaction into 
appropriate phases and to consider independently each phase. The mech- 
anism includes a free radical branching step that is autocatalytically 
inhibited by the presence of bromide. Oscillations are sustained by 
the alternation of the inhibition as the bromide concentration changes. 

Despite the elegant mechanism, it is not possible to deal fully 
quantitatively with the Belousov reaction because of the insolvable form 
of the general rate expression. Approximate rate expressions can be 
derived for specific portions of a given cycle. 11 Attempts have been 
made to computer simulate several proposed mechanisms that could be 
found in any oscillating reacting. 14 By supplying rate constants and 
initial concentrations, it is possible to establish whether a particu- 
lar mechanism generates oscillations. 10 Computer and mathematical 
methods for treating oscillating reactions lag behind the level of com- 
plexity required to deal with actual laboratory results. 

Bray Reaction 

In 19^1 Bray reported that a mixture of hydrogen peroxide, potas- 
sium iodate, and sulfuric acid in the appropriate concentrations showed 
oscillations in the rate of oxygen evolution. 15 Oscillations in the 
concentrations of iodine, iodide, and hydrogen ion have been identi- 
fied. 16 The Bray reaction appears to involve the following reactions: 17 

?HI0 3 + 5H 2 2 -> I 2 + 50 2 + 6H 2 

I 2 + 5H 2 2 -> PHIO3 + 4H 2 

The mechanism is not known, but it has been proposed to involve a radi- 
cal chain autocatalytic step that is activated and then inhibited to 
produce oscillations. 17 At present, the identity of the intermediates 
is an open question whose answer should allow an explanation for the 
periodic fluctuations. 

General Conclusion 

Free radicals are probably involved in many of the oscillating 
reactions in the autocatalytic step. 18 In addition to the Bray and 
Belousov reactions, free radicals are probably significant in the 
little studied oscillating dithionite decomposition reaction. 19 There 
are probably many more oscillating reactions to be discovered. The 
lack of interest in oscillating reactions and the need for instrumen- 
tation capable of continuous monitering of concentrations explains why 
few systems are known. Biological enzyme systems, which abound with 
feedback controls (allosteric enzymes), are the most likely class of 
reactions to have oscillations. 20 It should eventually be feasible to 
design inorganic oscillation reactions. 5 



-35- 

The knowledge gained from the study of oscillating systems 
should "be of significance in understanding biological periodicity. 
Applications for oscillating reactions in industry and in further 
academic research remain to be defined. 

References 

Nicolis and J. Port now, Chem. Rev., 73, 365 (1973). 

Pye and B. Chance, Pro. Nat. Acad. Sci.j J35, 888 (1968). 

Lewis and G. Von Elbe, "Combustion, Flames, and Explosions 

Gases," Academic Press, New York, 1951, pp. 133-138, 177. 

N. Noyes, R. J. Field, and E. KBrtts, J. Amer. Chem. Soc, 
94, 139^ (197?). 

Degn, J. Chem. Ed., 49, 30? (1972). 

F. Gray, Trans. Faraday Soc, §6, 363 (1970). 

Prigogin, R. Lefever, A. Goldheter, and M. Hershkowitz- 

ifman, Nature, 229 , 913 (1969). 

J. Field, J. Chem. Ed., 49, 308 (1972). 

J. Lotka, J. Amer. Chem. Soc, 42, 1595 (1920). 

Higgins, Ind. and Eng. Chem., 59 (5), 19 (1967). 

J. Field, E. Kbrbs, and R. M. Noyes, 94, 8649 (1972). 

Kopell and L. N. Howard, Science, 180, 1171 (1973). 

J. Kasperk and T. C. Bruice, Inorg. Chem., 10, 382 (1971). 

Lindblad and H. Degn, Acta. Chem. Scand., 21. 791 (1967). 

C. Bray, J. Amer. Chem. Soc, _43, 126? (19"2T). 

H. Woodson and H. A. Liebhafsky, Nature, 224 , 690 (1969). 

Degn, Acta. Chem. Scand., 21, 1056 (1967). 

L. Noyes, Abstract of Papers, American Chemical Society 
165th National Meeting, 1973, Phys. 76. 

19. R. G. Rinker, S. Lynn, D. M. Mason, and W. H. Corcoran, 
Ind. Eng. Chem. Fundam. , 4, 282 (1965). 

20. K. Pye, Cand. J. Botany, ¥7, 271 (1969). 



1. 


G. 


2. 


K. 


3. 


B. 




of 


4. 


R. 




94 


5. 


H. 


6. 


B. 


7. 


T. 


8. 


iva 
R. 


9. 


A. 


10. 


J. 


11. 


R. 


12. 


N. 


13. 


G. 


14. 


P. 


15. 


W. 


16. 


J. 


17. 


H. 


18. 


R.. 



THE SPIN LABEL TECHNIQUE AND ITS ESR APPLICATIONS 

Marinda P. Li April 2, 19Jk 

Introduction 1 " 8 

A spin label is a synthetic, paramagnetic, site-directed, 
organic free radical. The first spin label back in 19^5 was the 
cation radical of chloropromazine, 17 but the most successful com- 
pounds used as spin labels to date are nitroxide free radicals. The 
technique consists of introducing a spin label into the system of 
interest and observing the free radical by electron spin resonance 
spectroscopy. Information available from the esr spectrum includes 
polarity of environment, orientation and motion of the spin label, 
and proximity of two or more spin labels. Since the pioneering 
papers of H. M. McConnell in 1965* the spin label literature has 
grown to a prodigious size covering a wide range of applications to 
biological problems. 

Theory 2 * 4 * 6 * T 

A very important feature of nitroxide spin labels is the com- 
parative simplicity and high accuracy of the spin quantum mechanic s 
that relates the physical state of the label to its paramagnetic 
resonance spectrum. 2 The electron spin resonance spectrum of a 
typical nitroxide in solution at room temperature is simple, con- 
sisting of three sharp lines produced by nitrogen hyperfine inter- 
action. The g tensor and nitrogen hyperfine coupling tensor are 
solvent dependent 9 * 10 and, thus, solvent effects can be used to probe 
the hydrophobic or hydrophilic nature of the environment around the 
label. The widths of the individual esr lines 10 * 11 are markedly de- 
pendent upon the rotational mobility of the nitroxide. The general 
dependence of resonance spectra on the molecular motion of labels 
makes these paramagnetic molecules delicate indicators of confor- 
mation changes in macromolecules. 2 

Dinitroxide spin labels allow the study of certain intramolecular 
interactions, since the interaction between the two unpaired elec- 
trons of a biradical is a strong function of the conformation of the 
biradical. Although the electron-electron exchange parameter J is 
not easily related quantitatively to molecular geometry, the obser- 
vation that an esr spectrum can be grossly altered by changes in 
structure or conformation renders J a potentially important parameter 
in spin labeling. 12 " 15 The electron-electron dipole term can also, 
in principle, provide quantitative distance and conformational infor- 
mation in dinitroxide spin labeling experiments. The point dipole 
approximation has been used to relate the dipolar splitting to the 
distance between two nitroxides. 16 

Applications 

It is convenient to group spin labels into three classes: 
(l) covalently binding spin labels, (2) noncovalently binding spin 
labels, and (3) spin- labeled analogs of biological molecules. 



-37- 

McConnell and co-workers 1 * 18 " 22 took the approach of covalent 
attachment of an iodoacetamide spin label to alkylate cysteine 3-93 
of hemoglobin in their studies of the cooperativity of oxygen binding 
in hemoglobin. From several independent esr experiments, they ob- 
tained evidence that the protein structure in the vicinity of the spin 
label attached to P-93 depended on the state of oxygenation of the a- 
chains, thus, suggesting a- 8 subunit interactions. 

Wood and co-workers 23 * 24 synthesized an active spin-labeled co- 
binamide in order to elucidate the role of vitamin B i2 as a coenzyme. 
Esr was used to study the kinetics of the homolytic cleavage of the 
Co-C bond for these spin- labeled analogs of vitamin B i2 . 

One of the most popular applications of spin labels over the last 
few years has been their use as probes of biomolecular structure. 4 * 6 * 7 * 15 
Hubbell and McConnell 25 undertook an exploratory study of the structure 
of a number of biological membranes including the excitable membranes 
of nerve and muscle. They allowed a small unreactive nitroxide spin 
label to diffuse into several membranous systems. Interpretation and 
comparison of the characteristic esr spectra obtained for this para- 
magnetic probe in various systems led these workers to conclude that 
excitable membranes must contain either phospholipid bilayers similar 
to those present in phospholipid vesicles or other components with very 
similar hydrophobic regions. 

Conclusion 

As in all approaches involving molecular modification, one must 
consider to what degree a spin label alters the system. Although there 
are many examples of labeled systems maintaining the functions of the 
unmodified system, each system behaves differently. Thus, care should 
be exercised in interpreting spin label spectra. However, by careful 
choice of appropriate spin labels and labeling sites, it should be 
possible to study almost any desired system without significant 
structural perturbation. 

One of the potentially most significant aspects of spin label 
spectra is that they can be observed in many situations --in protein 
single crystals, in solution, and in more complex structures such as 
membranes. The addition of this versatile technique to the battery 
of spectroscopic tools available to the biologist promises to add in- 
formation in appropriately selected systems, which can be combined 
with information accumulating from other techniques. 



REFERENCES 
Reviews 

1. H. M. McConnell, Annu. Rev. Biochem ., 40, 227 (1971). 

2. H. M. McConnell and B. G. McFarland, Quart . Re v . Bi ophys . , 3, 
91 (1970). 

3. C. L. Hamilton and H. M. McConnell, Structural Chemistry and 
Molecular Biology , ed., A. Rich, N. Davidson, p. 115, 

W. H. Freeman and Co., San Francisco (1968). 

4. P. Jost, A. S. Waggoner, and 0. H. Griffith, Structure and 
Function of Biological Membranes , ed., L. I. Rothfield, p. 83, 
Academic Press, New York (1971). 

5. 0. H. Griffith and A. S. Waggoner, Acct. Chem. Res ., 2, 17 (1969). 

6. I. C. P. Smith, Biological Applications of Electron Spin 
Resonance Spectroscopy , ed. , J. R. Bolton, D. Borg, H. Swartz, 
Chapt. 11, John Wiley and Sons, New York (1970). 

7. 0. H. Griffith, L. J. Libertini, and G. B. Birrell, 
J. Phys. Chem ., 75, 3417 (1971). 

8. E. Rosantzev, Free Nitroxyl Radicals , Plenum, New York (1970). 

Theory 

9. R. Briere et. al., Bull. Soc. Chim. Fr ., 3273 (1965). 

10. T. J. Stone et. al., Proc. Natl, Acad. Sci ., 54, 1010 (1965). 

11. A. Carrington and H. C. Longuet-Higgins, Mol. Phys ., 5. 447 (1962). 

12. G. Luckhurst and G. Pedulli, J. Amer. Chem. Soc , 92, 4738 (1970). 

13. R. Briere et. al., Bull. Soc. Chim. Fr ., 5290Tl963T. 

14. H. Lemaire, A. Rassat, and P. Rey, Bull. Soc. Chim. Fr . , 886 (1968). 

Applications 

15. M. Calvin et. al., Proc. Natl. Acad. Sci ., 65 , 1 (1969). 

16. Z. Ciecierska-Tworek, S. P. Van, and 0. H. Griffith, 
J. Molec. Struct ., 16, 139 (1973), 

17. S. Ohnishi and H. M. McConnell, J„ Amer. Chem. Soc , 87, 2293 (1965). 

18. S. Ogawa and H. M. McConnell, Proc. Natl. Acad. Sci ., "38, 19 (1967). 

19. S. Ogawa, H. M. McConnell, and A. Horwitz, Proc. Natl.Tcad. Sci ., 
61, 401 (1968). 

20. H. M. McConnell, S. Ogawa, and A. Horwitz, Nature , 220 , 787 (1968). 

21. J. C. A. Boeyens and H. M. McConnell, Proc. Natl. Acad. Sci. , 
56, 22 (1966). 

22. J. K. Moffat, J. Mol. Biol ., 55, 135 (1971). 

23. T. Buckman, F. S. Kennedy, and J. M. Wood, Biochem . , 8, 4437 (1969). 

24. J. M. Wood et. al., Biochem . , 10, 3428 (197TJ7 

25. W. L. Hubbell and H. M. McConnell, Proc Natl, Acad. Sci ., 6l, 
12 (1968). — 



-39- 

OBSERVING THE BEHAVIOR OF TRANSIENT CHEMICAL SPECIES 
Mitch Hoselton April 11, 1974 

Short-lived species, be they chemical intermediates or excited 
electronic states, may be studied by a variety of methods. Each of 
these methods has its own characteristic time domain. The limitations 
on the time domain may arise from either the uncertainty principle or 
may simply be due to experimental difficulties of the more practical 
and therefore avoidable sort. 

Flash photolysis lies in this latter category. The problem in the 
past has always been to generate short enough pulses with sufficient 
intensity to observe the kinetically interesting, but short-lived 
species. New developments in laser technology have to some degree al- 
leviated problems of this sort and it is now possible to generate pulses 
and observe spectral changes whose lifetimes are on the order of 
10" 11 sec. 

Flash Photolysis 

Here, the source of radiant energy is a gas discharge tube which 
is "fired" by shorting a condenser across the tube. 1 The wide choice 
of spectral regions and intensities accessible by using different tubes 
gives this technique good flexibility. The fundamental drawback has 
been the duration of the pulses. So-called "de convolution" techniques 
allow the experimentalist to calculate lifetimes which are smaller than 
the duration of the flashes, 2 but this seems to be at best a second- 
order guess. Depending on the configuration of the apparatus, lifetimes 
on the order of 10" T sec may be observed directly. 

Q-Switched Lasers 

Lasers offer several advantages over the conventional flash lamp. 
Better time resolution (10~ 9 sec vs 10" 7 sec) is one, and they are also 
more efficient in delivering radiant energy to the sample. Thus, while 
flash lamps may generate hundreds or thousands of Joules of energy, 
much of this is wasted and a laser, though generating only 2-10 J, can 
produce a measurable effect by focusing all of it into the sample. The 
use of amplifying rods exterior to the laser is useful in these appli- 
cations for providing additional power and transient concentrations. 
The use of frequency doubling 3 provides some spectral flexibility, while 
optical delays and non- linear optical phenomena (white light 4 gener- 
ation) allow use of the laser for the dual role as source and monitor 
of the transient behavior. 

Recent articles by Sutin 5 detailing the use of these lasers as 
T-jump sources extends the flexibility for these particular systems. 
Thus, when the sample won't absorb at available frequencies, a solvent 
which does might be found. 



-40- 



Mode- Locked Lasers 



It is now possible to extend the range of measurable lifetimes 
down to the pico-second (10 12 sec) range. 8 The power in such pulses 
is still very high (power=energy per unit time) even though the abso- 
lute energies are relatively low. The primary difference between this 
method and the two above is that the output consists of a series of 
pulses rather than just one pulse, so that experimental results are the 
sum of a number of separate experiments (assuming the system has re- 
covered completely from one pulse before the next pulse arrives). 
Variation of the pulse separation is possible over the range from 2-10 
nsec in practice. 

Applications 

Chlorophyll systems have now been studied by these techniques. 
Classical flash photolysis techniques are inadequate to resolve the 
fluorescence decay time. 7 " More sophisticated techniques using spark 
gap sources and photon counting equipment have been used to estimate 
lifetimes of from 0.6 to 1.6 n-sec without much apparent consistency. 2 
The most recent measurement using a mode- locked laser, however, shows 
that there are probably two different fluorescing species and that the 
1/e decay time is only 350 p-sec. 8 

Using a Q- switched laser system, various groups have examined the 
equilibrium 

NiL 2+ + 2H 2 * NiL(H 2 0)| + 

where L = N, N l -bis(2-aminoethyl)-l,3-propanediamine. 9 ' 10 Changing the 
incident frequency or the solvent made it possible to check the rela- 
tive effects of the T-jump and photochemical perturbations. Even though 
these two perturbations push the equilibrium in opposite directions, the 
kinetic lifetimes were in close agreement, demonstrating the general 
validity of the interpretation of the photochemically obtained results 
in terms of the reactions of ground-state species. 

The geometry of the photochemically generated metal pentacarbonyls 
of Cr, Mo and W have also been examined. 11 *" 13 Early reports of the 
observation of the D 3 ft form presently appear to be in error. 14 The 
absorption spectrum of the intermediate Cr(C0) 5 is similar to that re- 
ported for Cr(C0) 5 THP, suggesting that the C 4v form is favored. This 
may be due to the ability of Cr(C0) 5 to scavenge trace impurities 13 in 
hydrocarbon solvents in order to satisfy its requirement for six- 
coordination. This high affinity for an octahedral structure may be 
related to the undesirability of the triplet ground state required in 
a l6e" D 3 n system. 12 

The reactions of I 2 in the presence of iodide in aqueous solution 
have also been studied using Q- switched lasers in the T-jump 15 and 
photochemical excitation modes. 16 " 18 In the_former case, the very rapid 
equilibrium rate constants between i 2 and I 3 ~ have been measured. In 
the latter case, under the highest time resolution used, no direct 
evidence of the proposed I- complex was observed, although good evi- 
dence for the existence of an I 2 intermediate was presented. 



-4l- 



REFERENCES 



1. a) Physical Methods of Chemistry , A. Weissberger and 
B. W. Rossiter, eds., Wiley and Sons, New York (1972), 
pt. IIIB, p. 521. 

b) F. N. Willets, Prog. React. Kin., 6(l), 51 (1971). 

2. W. J. Nicholson and J. I. Fortoul, Biochem. Biophys. Acta, 
143, 577 (1967). 

3. R. L. Bulton and B. Watson, J. Chem. Phys., 58, 1778 (1973). 

4. C. A. E. 0. Varma and P. M. Rentzepis, J. Chem. Phys., 58, 
5237 (1973). 

5. J. V. Beitz, G. W. Flynn, D. H. Turner, and N. Sutin, 
J. Amer. Chem. Soc., 92, 4 130 (1970). 

6. M. A. Duguay and J. W. Hanson, Optics Comm. , I, 254 (1969). 

7. W. L. Butler and K. H. Norris, Biochem. Biophys. Acta, 66 , 
72 (1963). 

8. M. Seibert, R. R. Alfano, and S. L. Shapiro, Biochem. Biophys. 
Acta, 292, 493 (1973). 

9. K. J. Ivin, R. Jamison, and J. J. McGarvey, J. Amer. Chem. Soc, 
94, 1763 (1972). 

10. TT7 Creutz and N. Sutin, J. Amer. Chem. Soc, 95, 7177 (1973). 

11. J. A. Mclntyre, J. Phys. Chem., 74, 2403 (19707. 

12. M. J. Boylan, P. S. Batterman, and A. Fullerton, 
J. Organomet. Chem., 31, C29 (1971). 

13. J. M. Kelly, H. Hermann, and E. K. von Gustorf, J. C. S. Chem. 
Comm., 1975 , 105. 

14. J. Nasielski, P. Kirsch, and L. Wilputte-Steinert, 
J. Organomet. Chem., 29, 269 (1971). 

15. D. H. Turner, G. N. Flynn, N. Sutin. and J. V 6 Beitz, 
J. Amer. Chem. Soc., 94, 1554 (1972). 

16. L. I. Grossweiner and M. S c Matheson, J. Phys. Chem., 6l, 
1089 (1958). 

17. N. K. Bridge, J. Chem. Phys., 2>2, 945 (i960). 

18. P. F. de Violet, R. Bonneau, and J. Joussot-Dubien, 
Mol. Photochem. , 5 S 6l (1973). 



-41- 

SECONDARY PROCESSES IN PHOTOELECTRON SPECTROSCOPY 

Edward J. Laskowski April 16, 1974 

Until recently, the main thrust of research activity in X-ray 
Photoelectron Spectroscopy (ESCA) has been in the correlation of 
binding energies of inner core electrons with various parameters in- 
dicative of charge density such as Mossbauer chemical shifts, 1 
oxidation states, 2 and the results of various molecular orbital cal- 
culations. 3 These studies have all been concerned with the principal 
ionization lines and have ignored the less easily detectable and 
identifiable satellite lines. 

The first step of the photoionization process is the creation of 
an inner shell hole by the interaction of the X-ray with the atomic 
electrons. Relaxation of outer orbitals takes place on a time scale 
which is long, compared with the photoejection, so that it is valid to 
consider the final state of the system as containing an inner orbital 
photohole. However, the electrons in the other shells can respond 
quickly to the inner shell hole by adiabatic orbital shrinkage. To a 
first approximation, the orbitals may be considered "frozen", and this 
effect neglected. 4 Under the frozen- orbital approximation, energy 
transitions can arise only from electric monopole interactions and re- 
sult in single electron ionizations. As a result, only transitions 
from well defined energy levels can occur, and hence the assignment of 
peaks to these levels. Within this approximation, additional structure 
can be introduced by two processes. These are: (l) Exchange splitting 
in open shell configurations, in which the final states differ in 
energy due to differing spin multiplicity, 5 and (2) Multiplet splitting 
in which configuration interaction between a core level and a partially 
filled valence level gives rise to several energetically different 
states. 6 Additional peaks can be observed due to the less interesting 
effects of nonmonochromatic excitation and impurities, especially 
metal oxides which form on the surface of the sample during irradiation. 

Double electron excitation becomes allowed under the sudden 
approximation, 7 ' 8 in which the orbitals are no longer frozen. In this 
case, the photoionization is considered as a two step process in which 
the photoionization occurs, and then simultaneous excitation of a 
second electron results due to the change in the screening, together 
with the relaxation of the remaining electrons from neutral atom 
orbitals to ionized atom orbitals. If the process results in ionization 
of two electrons, it is referred to as shake-off, whereas if it re- 
sults in the ionization of one electron and the promotion of a second 
electron to an excited state, it is referred to as shake-up. 

A third two-electron process which is possible is Auger electron 
emission. 9 This is a radiationless process in which an electron oc- 
cupying a higher energy orbital relaxes to fill the photohole formed 
by the primary photoionization. Simultaneously with this relaxation, 
there is an ejection of a second electron. Although this gives rise 
to fairly intense peaks, this is not frequently studied due to the 
restrictions on the energy of the second ejected electron. 10 



-42- 

ESCA studies of various transition metal carbonyls have been 
undertaken. 11 These spectra show distinct satellites, separated by 
5-6 eV from the normal ionization peaks for C(ls) and 0(ls) elec- 
trons. The metal center exhibits broadening of the principle ion- 
ization peak. These satellites have been attributed to shake-up 
due to charge transfer excitations, and indeed, the energy separations 
for the C(ls) and 0(ls) satellites correspond well with the trans- 
ition lAxp. to d x T! , which is the dominant feature in the UV ab- 
sorption spectra or these complexes. From consideration of the 
differences between the UV dipole excitation and the ESCA monopole 
excitation, 12 the agreement between these two energies seems fortui- 
tous and leads to the interpretation that the initial and final states 
are almost equally influenced by core ionization. The intensity vari- 
ation suggests that the two electron excitation process is favored 
when the positive inner hole is located on the acceptor part of the 
molecule, since electron flow during the charge transfer process is 
favored towards this center. Likewise, the positive hole on the donor 
inhibits the charge transfer process, and hence the satellite is of 
low intensity at the metal site. 13 Further evidence for this type of 
transition was given in a systematic study of substituted chromium 
carbonyls of the type Cr(CO) 5 X, which revealed a reduction of the 
intensity of the oxygen satellite relative to the 0(ls) ionization 
with increasing chromium- carbon bond strength. 14 This is in agreement 
with a charge transfer excitation in the shake-up mechanism, as this 
should be inhibited by the presence of a strong electron donor on the 
metal center. 

Recent studies of the satellite structure of 46 copper compounds 15 
and 70 nickel compounds 16 have revealed features which are peculiar 
to the electronic configuration of the metal. It has been found that 
only paramagnetic metal centers give rise to satellite structure on 
the metal 2p signals. The 2p signals are split in all cases into a 
2P3/2 component and a 2p x / 2 component due to spin-orbit coupling. 
Three specific patterns of satellites arise in the paramagnetic centers: 
Type A, in which both 2p levels show two satellites; Type B, in which 
one 2p level shows two satellites and the other 2p level shows only 
one; and Type C, in which there is only one satellite associated with 
each 2p signal. Although the specific origin of these satellites is 
as yet uncertain, it has been suggested that the high energy satellite 
may be due to a shake-up transition localized on the metal center with 
excitations of the type 3d to 4s or 3d to 4p, while the lower energy 
satellite may be due to a shake-up transition of the type 3d to ligand 
antibonding orbital. 

Possible application of these observations to structural chemistry 
has already been shown. In the case of (K 2 Ni(CN) 3 ) 2 > which is known 
to have Ni(l) paramagnetic centers, no satellites were observed, indi- 
cating that the metals were diamagnetic. X-ray diffraction studies 
have shown the compound to be dimeric and have a short metal to metal 
distance, indicative of a metal-metal bond, which is consistent with 
the ESCA data. 16 The electronic structure of nickel dithiolate com- 
pounds was also clarified by the observation of ESCA satellites. 1 " 7 
These compounds were known to be paramagnetic, both in solution and 
in solid state. This was interpreted as indicating that the compounds 



-43- 

contained nickel (ill) ions. Molecular orbital calculations indi- 
cated that the unpaired electrons were localized mainly on the ligand, 
however. The absence of shake-up satellites confirmed the latter 
interpretation. 

Although the exact mechanism for multiplet splitting is still 
unclear, it appears that the observation of satellite lines will be 
useful in the interpretation of electronic structure, oxidation states, 
and stereochemistry. Hopefully, the problems with instrumental reso- 
lution can be solved, so that further progress can be made in this area, 

REFERENCES 

1. a. W. E. Swartz, P. H. Watts, Jr., E. R. Lippincott, J. C. Watts, 
J. E. Huheey, Inorg. Chem. , 11, 2632 (1972). 

b. M. Barber, P. Swift, D. Cunningham, M. J. Frazer, 
Chem. Commun. , 338 (1970). 

2. D. P. Murtha, R. A. Walton, Inorg. Chem., JL2, 368 (1973), and 
references therein. 

3. L. N. Kramer, M. P. Klein, Chem. Phys. Lett., 8, 183 (1971), 
and references therein. 

4. J. M. Hollander, W. L. Jolly, Accounts Chem. Res., 3, 193 (1970), 

5. G. K. Wertheim, A. Rosencwaig, R. L. Cohen, H. J. Guggenheim, 
Phys. Rev. Lett., 27, 505 (1971). 

6. C. S. Fadley, D. A. Shirley, Phys. Rev., A., 2, 1109 (1970). 

7. T. Aberg, Phys. Rev., 156, 35 (1967). 

8. M. 0. Krause, T. A. Carlson, R. D. Dismukes, Phys. Rev., 170 , 
37 (1968). 

9. D. M. Hercules, Anal. Chem.. 42, 20A (1970). 

10. C. D. Wagner, Anal. Chem., 44, 967 (1972). 

11. S. Pignataro, Z. Naturforschung, 27, 816 (1972). 

12. T. Novakov, R. Prins, Solid State Comm., 9, 1975 (1971). 

13. S. Pignataro, G. Distefano, J. Electron Spec. Relat. Phen. , 2, 
171 (1973). 

14. M. Barber. J. A. Connor, I. H. Hillier, Chem. Phys. Lett., 9$ 
570 (1971). 

15. D. C. Frost, A. Ishitani, C. A. McDowell, Mol, Phys., 24, 
861 (1972). — 

16. L. J. Matienzo, L. I. Yin, S. 0. Grim, W. E. Swartz, Jr., 
Inorg. Chem., 12, 2762 (1973). 

17. S. 0. Grim, L. J. Matienzo, W. E, Swartz, Jr., Inorg. Chem., 
13, 447 (1974). 



-45- 



MAGNETIC EXCHANGE INTERACTIONS IN DIMERIC Cu(ll) COMPLEXES. 

(Thesis Seminar) 
Elvira F. Hasty April 18, 1974 

In recent years, much attention has been given to the magnetic 
properties of binuclear transition metal complexes, in particular, 
the Cu(ll) complexes. 1 ' 2 These systems can have either a direct metal- 
metal interaction 3 ' 4 or, the most common, an exchange interaction 
through bridging groups. In most cases, the exchange interaction is 
antiferromagnetic (singlet ground state), 5 " 21 while a few compounds with 
a triplet ground state have also been observed. 15 > 22 ~ 2£ Usually, the 
bridging units are small molecules or ions, and the internuclear 
distances are not very large. Little esr work has been reported for 
these complexes, and the exchange mechanism has been treated in only a 
few of the systems. 15 * 17 * 24 * 25 

The objective of this work was to synthesize dimer complexes from 
binucleating ligands with aromatic systems acting as bridging moieties 
in order to determine the possibility of exchange interactions in such 
systems. The compounds were characterized by analyses, infrared and 
either molecular weight determinations or mass spectrometry. The mag- 
netic properties were studied by variable-temperature (4.2-290°K) 
magnetic susceptibility measurements and electron spin resonance. 

Two kinds of binuclear complexes were studied: 1. Jager-Type 
Complexes . These compounds were synthesized by a modified method of 
the reactions used by Jager 26 in the preparation of monomeric square- 
planar complexes. The ligands formed from condensation reactions of a 
6-diketone and the appropriate amine, either 1, 2, 4,5-tetraaminobenzene 
or 3,3' ,4,4' -tetraaminobiphenyl. 2. Schiff-Base Complexes . The 
ligands formed from condensation reactions of salicylaldehyde and the 
appropriate amine, either m-phenylenediamine, p-phenylenediamine 2T or 
2,2' ,6,6' -tetraaminobiphenyl. The Cu(ll) complexes for both types of 
compounds were prepared by reacting the ligand with Cu(OAc) 2*H 2 0. 

An intramolecular antiferromagnetic exchange interaction was 
found to be present in all of the Cu(ll) complexes except when the 
bridging group is 3,3' ,4,4 • -tetraaminobiphenyl. The strength of the 
exchange interaction was found to depend on such factors as type of 
orbital where the unpaired electron is present, internuclear distance 
and aromaticity of the bridging unit. The mechanism of exchange is 
explained in terms of derealization of the unpaired electron of the 
Cu(Il) ions into the aromatic bridging unit by means of overlap of 
the orbitals where the unpaired electron is present with the 
appropriate molecular orbitals of the ligand. 



-46- 

References 



1 
2 
3 

4 

5 

6 

7 

8 
9 

10 

11 
12 
13 
14 

15 
16 

17 
18 

19 
20 
21 

22 

23 

24 
25 

26 
27 



M. Kato, H. B. Jonassen and J. C. Fanning, Chem. Rev., 64, 99 (1964). 

W. E. Hatfield and R. Whyman, Transition Metal Chem., 5, 47 (1969). 

B. Bleaney and K. D. Bowers, Proc. Roy. Soc. (London), A214 , 

4511 (1952). 

R. L. Martin and H. Waterman, J. Chem. Soc., 2960 (1959). 

W. E. Hatfield and J. A. Crissman, Inorg. Nucl. Chem. Lett., k, 

731 (1968). 

W. E. Hatfield and G. W. Inman, Jr., Inorg. Chem., 8, 1377 (1969). 

L. K. Thompson, V. T. Chacko, J. A. Elvidge. A. B. P. Lever, and 

R. V. Parish, Can. J. Chem., 47, 4l4l (I9b9). 

H. Ojima and K. Yamada, Proc. Symp. Coord. Chem., 3rd, 1, 28l (1970). 

M. Kato, Y. Muto, H. B. Jonassen, K. Imaii and T. Tokii, 

Bull. Chem. Soc. Japan, 143, 1066 (1970). 

G. 0. Carlisle and W. E. Hatfield, Inorg. Nucl. Chem. Lett., 6, 

633 (1970). 

B. J. Cole and W. H. Brumage, J. Chem. Phys., 53, 4718 (1970). 

J. F. Villa and ¥. E. Hatfield, J. Amer. Chem. Soc., 93, 4o8l (1971). 
H. Okawa and S. Kida, Bull. Chem. Soc. Jap., _44, 1172"^"l97l) . 
S. Kokot, C. M. Harris and E. Sinn, Aust. J. Chem., 25, 45 (1972); 
earlier papers are referenced. 

D. Y. Jeter, D. L. Lewis, J. C. Hempel, D. J. Hodgson and 
W. E. Hatfield, Inorg. Chem., 11, 1958 (1972). 

W. E. Hatfield, Inorg. Chem., 11, 2l6 (1972); earlier papers 

are referenced. 

J. F. Villa and W. E. Hatfield, Inorg. Chem., 11, 1331 (1972). 

P. Singh, D. Y. Jeter, W. E. Hatfield and D. J. Hodgson, 

Inorg. Chem., 11, 1657 (1972). 

G. W. Inman, Jr., and W. E. Hatfield, Inorg. Chem., 11, 3085 (1972). 

J. F. Villa, Inorg. Chem., 12, 2054 (1973). 

K. Emerson, A. Emad, R. W. Brookes and R. L. Martin, 

Inorg. Chem., JL2, 978 (1973). 

J. A. Barnes, W. E. Hatfield and D. J. Hodgson, Chem. Comm. , 1593 

(1970). 

W. E. Hatfield, J. A. Barnes, D. Y. Jeter, R. Whyman and 

E. R. Jones, Jr., J. Amer. Chem. Soc, 92, 4982 (1970). 

W. E. Hatfield and J. F. Villa, Inorg. Chem., 10, 2038 (1971). 
J. A. Barnes, D. J. Hodgson and W. E. Hatfield, Inorg. Chem., 11 , 
144 (1972). 
Von L. Wolf and E. G. Jager, Z. Anorg. Allg. Chem., 346, 76 (1966). 

C. A. Bear, J. M. Waters and T. N. Waters, J. Chem. Soc. (A), 
2494 (1970). 



-47- 



BINUCLEAR COMPLEXES OF COBALT (II) 
Nelson B. O'Bryan April 19» 197^ 

The complex [a,a°-(ethylenedinitrilo)di-o-cresolato)] cobalt(II), 
Co(salen), has been extensively studied as an oxygen carrier* Like 
its copper analog, Co(salen) may also function as a bidentate ligand 
which in effect produces a binuclear complex with phenolic oxygen 
bridges. It has, beenpfoundpthat withpjri- or hexafluoroacetylacetone 
complexes of Ni , Cu , Zn , and Cd , Co(salen) does produce an 
adduct, the expected reaction being 

0* 





H ?°. r^ + ,/ m n, * L' Co wSn 



C=N' N=C X N 




Co(salen) R = CH-, CF 

2+ 2+ 
The product illustrated above is formed if M = Zn or Cd ; 

however, for M = Ni or Cu the final product is best described as 

an adduct_of M(salen) and Co(Facac) ? , (Facac «= fluorinated acetyl- 

acetonato") , i.e©, the two metals have exchanged coordination spheres. 

The fact that exchange has occurred in these cases is supported by 

magnetic susceptibilities and mass spectra. 

The rate of metal exchange which is quite rapid when hexafluoro- 
acetylacetone complexes are employed is significantly slower for the 
trifluoroacetylacetone analogs and does not occur at all for unsub- 
stituted acetylacetone complexes. Kinetic studies of the trifluoro- 
acetylacetone systems, while inconclusive in determining a rate 
expression or mechanism, do show that the reaction rate has a complex 
dependence on both reactant concentrations. 

Molecular weight determinations, visible and ESR spectra show 
that the binuclear species undergoes negligible dissociation in non- 
coordinating solvents such as methylene chloride. 

The crystal and molecular structure of the Cu(salen)Co(hfa) ? 
adduct has bee ^determined by X-ray diffraction (hfa = hexafluoro- 
acetylacetonato") . The + structure, similar to that shown in the above 
diagram, contains a Cu with square planar coordination and a Co 
with a distorted octahedral environment. The copper-cobalt distance 
is 3-06 A. 



-48- 



Solid state, room temperature magnetic susceptibilities indicate 
that when cobalt is in the salen moiety, its magnetic moment is 
approximately 2.8 BM, a high, but not uncommon moment for square 
planar Co • If the cobalt atom is octahedrally coordinated, it is 
high spin with a magnetic moment of ™ ^.8 BM, a characteristic value 
for this type of environment. 

The temperature dependence of the magnetic susceptibility of 
Ni(salen)Co(hfa) 2 # which has only one paramagnetic center, was found 
to be quite ordinary over the temperature range l6~290°Ki X = 
3-13/(T + 16) • Cu( salen)Co(hfa) 2 gave a non-linear Curie-Weiss plot. 
Considerations of the susceptibilities indicate that some antiferro- 
magnetic coupling may be present between cobalt and copper. However, 
low temperature ESR data is in apparent conflict with this hypothesis© 



-49- 



BONDING IN TRANSITION METAL BINARY HYDRIDES 

Ma'mun Absi-Halabi April 23, 197*1 

INTRODUCTION 1 > 2 > 3 

Binary metallic hydrides are a wide class of metal- hydrogen 
compounds formed by the reaction of highly purified hydrogen with 
some of the transition metals and rare- earth elements in a pure 
form. The range of this class along with the other classes of hy- 
drides known are shown in the accompanying periodic table. 

Metallic hydrides exist in the form of brittle crystals, but 
they exhibit properties similar to those of metals, 3 namely, 
thermal conductivity, electrical resistance, hardness and luster. 
They have a variety of applications in industry and, on a purely 
scientific level, they have unique properties compared to other 
hydrogen compounds. 

This seminar will be limited to the nature of the bonding in 
this class, which has been a subject of controversy for a long time. 
The oldest model proposed was the solution of a gas in a solid where 
the hydrogen atoms occupy the interstitial sites of the metallic 
crystal without any form of electronic interaction with the metal 
atoms. This model had been rejected on the basis of the changes in 
crystal structure observed in some cases and the thermodynamic 
data obtained. 

Since the development of atomic theory, two models of bonding 
have been mainly used in discussing the experimental results of 
metallic hydrides: 

a) The Protonic Model 1 < 2 * 4 

The hydrogen Is orbital is assumed to be of higher energy than 
that of the Fermi surface of the metal. Consequently, the hydrogen 
atoms donate their electrons to the metallic bands and attain a (+l) 
oxidation state. 

This model was able to account for the properties of palladium 
hydride, PdHo.e* with some success. 

b) The Anionic Model 1 * 2 ' 4 

The hydrogen Is orbital is assumed to be of lower energy than 
that of the Fermi surface of the metal, thus it is assumed that it 
has a (-1) oxidation state. 

This model was more widely accepted and was capable of account- 
ing for a greater variety of the metallic hydride properties. Two 
other models have also been reported, the covalent bond model 2 and 
the intermediary model, 4 but they were of more limited use. 



-50- 



AUGMENTED PLANE WAVE (APW) CALCULATIONS 



5-9 



Recently, solid state quantum mechanical calculations were 
carried out on some transition metal hydride phases. 5 " 8 The APW 
method 9 used for the calculations is characterized by assuming an 
ideal potential, the muffin-tin potential, which is periodic and 
consists of two parts; a spherically symmetrical part that extends 
to a distance r from a nucleus, and a zero potential part throughout 
the region, Ti-T 2 t between two spherical potentials of any two ad- 
jacent nuclei. The potentials were derived from Hartree-Fock-Slater 
coulomb potentials and charge densities of the hydrogen and the metal 
atoms. The crystal parameters used were those obtained from neutron 
diffraction and x-ray crystallography. 

The results of these calculations show that the energy bands 
for the monohydrides are lowered considerably with respect to those 
of the metal. No new bands appear below the Fermi surface, although 
some states of the metal drop below it upon formation of the hydride 
phase. The lowest band of the hydride is a hybrid of the metal and 
the hydrogen s orbitals. Charge density calculations indicate that 
there exists a slight increase in electron density on the hydrogen of 
palladium hydride compared to the free hydrogen. 

For the dihydrides, a new band, which is the antibonding combin- 
ation of the two hydrogen Is orbitals, appears below the Fermi sur- 
face. In addition, states that have the proper symmetry to interact 
with the hydrogen orbital are also lowered. The hydrogen atoms in 
this case are in a partially negative oxidation state. 

The trihydrides treatment also shows that the hydrogen is in a 
partially negative oxidation state. 

CONCLUSION 

Although these calculations provide somewhat quantitative re- 
sults in a unified way, more work should be done in order to resolve 
the remaining ambiguities and to provide a method for interpreting 
various trends, 

A possible relationship between the electronic interactions in 
metallic hydrides and those in some other closely related areas, 
namely, catalytic hydrogenation and complex metal hydrides, is likely. 
This study of metal hydrides is an effort for the purpose of finding 
such a relationship that might lead to a better understanding 
of catalysis. 



-51- 

REFERENCES 

1. T. R. P. Gibb, Jr., Prog. Inorg. Chem . , 3, 315 (1962). 

2. G. G. Libowitz, The Solid-State Chemistry of Binary 

Metal Hydrides , W. A. Benjamin, Inc. (1965). 

3. W. M. Mueller, J. P. Blackledge and G. G. Libowitz, 

Metal Hydrides , Ch. 1 and Ch. 12, Academic Press (1968). 

4. B, Stalinski, Berichte der Bunsen-Gesellschaft, 76, 724 (1972). 

5. A. C. Switendick, Solid State Comm. , 8, 1463 (1970). 

6. A, C, Switendick, Int. J. Quan. Chem. , No. 5, 459 (1971). 

7. D. E. Eastman, J. K. Cashion and A. C. Switendick, 

Phys. Rev. Let., 27, 35 (1971). 
8 f A. C, Switendick, Berichte der Bunsen-Gesellschaft, 

76, 535 (1972), 
9, L, F. Mattheiss, J. H. Wood, and A. C. Switendick, 

Methods in Computational Physics, 8, 64 (1968). 

GENERAL REFERENCES ON SOLID STATE QUANTUM MECHANICS 

G. F. Koster, Space Groups and their Representation, 

Solid State Physics , 5, 174 (1957). 
J, Callaway, Electron Energy Bands in Solids, 

Solid State Physics , 7, 99 (1958). 
A. P. Cracknell and K. C. Wong, The Fermi Surface , Ch. 1, 

Clarendon Press, Oxford (1973). 
J. C. Slater, Energy Bands and the Theory of Solids , 

Methods in Computational Physics, 8, 1 (1968), 



-52- 



-Soline- 



-Covalent- 



,A 


1 
II A 








-Metallic 














III B 


1 

IV B | 


~ 














Scries 






















of Boron 




UN 


(BeH,)„ 
Covalent 
















Hydrides 
















Complex 
Covalent 


















Series 


NaH 


MgHj 


III A 


IV A 


V A 


VI A 






IB 


II B 


(AIH 3 )„ 


of S. 
Hydrides 






VII A/ ' ' ' v III A \ 






KH 


CoHj 


ScH, 


T,H 2 
(cubic 
and 
totrog ) 


VH 
VHj 


CrH 
CrHj 


Mn 


Fe 


Co 


NiH 


CuH 


(ZnH 2 ) n 


(GoHj)„ 


Series 
of Ge 
Hydrides 








ZrHj 






















RbH 


S'H : 


YH 2 
YH 3 


(cubic 
and 
tetrog ) 


NbH 
NbH 2 


Mo 


Tc 


Ru 


Rh 


PdH 


A9 


(CdHj)„ 


(lnH)„ 
(l"H 3 ) n 


SnH 4 
Sn 2 H 6 






See 


HfHj 






















CsH 


BoH, 


Rare 
Earth 
Series 


(cubic 

and 

tetrag.) 


Tah 


W 


Re 


Os 


Ir 


Pt 


Au 


(HgH 2 ) n 


<TIH) n 
(TIH 3 )„ 


PbH 4 











Pm 


Sr 


.H, 




GdHj 


TbHj 
TbH 3 


DyH 2 


HoH 2 


ErHj 


TmHj 


YbHj 


UHj 


toH, , 


CM, j 


PrHj 3 


NdH, j 


? 


Sr 


,H 3 


EuHj 


GdHj 


DyH 3 


H0H3 


ErH 3 


TmH 3 


YbH 3 (?) 


UH 3 



AcHj 


ThH, 
T^H.j 


PoH 3 


UH 3 


NpH 2 
NpHJ 


PuHj 
PuHj 


AmHj 
AmHj (') 



IliiiaiA hydrides lormcd bv metals in tlic periodic labltv 



-53- 

CALORIMETRIC STUDIES OP LEWIS ACID-BASE INTERACTIONS 
R, Martin Guidry April 2k, 197^ 

Intr oduc tion 

—M»— — ^"^- ill" "I — Mi l — OWI 

The principal reason for a chemist to carry out structural studies 
is the potential importance of this information in understanding 
chemical reactions and chemical reactivity. Although techniques avail- 
able for structural determination have improved considerably during 
the past decade, there has been few advances made during the past ten 
years in our understanding of chemical reactions and reactivity, A 
primary source of difficulty is the extreme complexity of typical 
chemical reactions-i-there are numerous energy terms contributing to 
chemical reactions which are not completely understood on the basis of 
the classical or quantum mechanical model. It is with this in mind 
that we have set out to study some of these energy terms. The initial 
step in this direction is the understanding of changes in the internal 
energy accompanying a very simple chemical reaction. The hope is that 
success will encourage study on more complex systems. 

Discussion 

A very simple reaction which is fundamental to a great number of 
complex reactions is the formation of addition compounds by Lewis 
acids and bases, 

A + :B ^ A:B (l) 

Over the past decade, a great deal of success has resulted from 
fitting the enthalpies of adduct formation (-AH) for a large number 
of Lewis acid-base interactions to an empirical four-parameter 
equation 1 * 3 -. -commonly called the E and C equation. 

-AH - E A E B + C A C B (2) 

E A and C« are empirical parameters for the Lewis acid and E Q and C R 
are the corresponding parameters for the Lewis base involvea in the 
reaction. By using the E and C equation, it is now possible to pre- 
dict the enthalpies of over 1500 Lewis acid-base interactions—many 
of which are extremely difficult to measure experimentally. 

When an attempt was made to obtain E A and C A parameters for the 
hydrogen bonding acid l,l,l,3,3,3-hexafluoro-2-propanol (HFIP) by 
computer- fitting calorimetrically measured enthalpies of interaction 
of HFIP with various Lewis bases, the fit was much poorer than one 
would anticipate. Two possible sources of the poor fit were a steric 
hindrance problem between one or more of the adducts or some constant 
contribution to all the measured enthalpies. The steric hindrance 
problem appeared improbable for an acid as small as a proton. The 
difficulty was believed due to the demonstrated 3 intramolecular 
hydrogen bonding in HFIP between the hydroxyl proton and -CF 3 fluorine. 
If this intramolecular bond has to be completely broken during the 
course of formation of the intermolecular Lewis acid-base bond, then 
this would comprise a constant contribution to the enthalpy. One 



-54- 

should then be able to subtract a constant amount from each of the 
experimentally measured enthalpies (which consist of two contri- 
butions—the breaking of the intramolecular bond and the formation 
of the Lewis acid-base bond) to obtain the enthalpy of the Lewis 
acid-base bond. This constant amount would represent the enthalpy 
of the intramolecular hydrogen bonding in HFIP. When this procedure 
was attempted, a best- fit of the E and C equation to the experi- 
mentally measured enthalpies was obtained when the constant amount 
was -1.1 kcal mole -1 . In order to simplify the procedure and obtain 
error limits for the value of the constant contribution, the E and C 
equation was modified to incorporate the constant contribution (w) 
and a computer program was written to solve for the three unknown 
parameters (E«, C« and W), 

-AH = E A E B + G A C B + W (3) 

Upon solving Equation 3 for the HFIP-Lewis base systems, it was 
found that the enthalpy of the intramolecular hydrogen bonding is 
-1.1 +0.3 kcal mole"" 1 . 

This procedure was then checked on several systems where the 
constant contribution can be determined independently. The procedure 
is invaluable for incorporating into the E and G scheme systems in 
which there is intramolecular association. It also has the potential 
of extending this treatment to dimers which must be dissociated to 
form adducts and whose enthalpy of dissociation is unknown. Dis- 
placement reactions also fall in the category of reactions containing 
a constant contribution to the enthalpy. 

Recently a model (called the elimination of solvation procedure) 
has been proposed 4 for estimating solvation contributions to enthalpy 
data in weakly basic, nonpolar and weakly polar solvents. The 
initial tests 4 "" 6 on this model, although encouraging, were incon- 
clusive because the range of solvent polarity was very limited. In 
this investigation, the model was put to a very rigorous test in which 
the range of solvent polarity was extended to include the very polar 
solvent nitrobenzene. Furthermore, the strong, hydrogen-bonding acid 
l,l,l,3,3,3-hexafluoro-2-propanol was employed along with an expanded 
range of Lewis bases. The data collected also permitted a test of a 
solvent transfer model proposed by Christian, et al. r After putting 
both models through the very rigorous tests, it was found that the 
elimination of solvation procedure model consistently predicted the 
correct results while Christian's model was less successful. 



-55- 



References 

1, R. S. Drago and B. B. Wayland, J. Amer . Chem , Soc t , 87, 
3571 (1965), 

2, R f S. Drago, G. C, Vogel and T. E. Needham, ibid., 93, 
6014 (1971). 

3, J, Murto and A, Kivinen, Suom , Kemlstilehti B, kO, 
li (1967) and references therein. 

4, R, S. Drago, M t S, Nozari and G. C. Vogel, J, Amer , Chem , Soc . , 
2Jt, 90 (1972), 

5, M, S. Nozari and R. S, Drago, Inorg . Chem . , 11 , 280 (1972), 

6, M, S, Nozari and R, S. Drago, J. Amer . Chem . Soc ., 9^, 
6877 (1972), -"" 

7, S, D, Christian, J, Phys . Chem., 70, 3376 (1966), 

1 



-do- 



ELECTRONIC INTERACTIONS BETWEEN PAIRS OF METAL IONS BRIDGED BY SMALL ANIONS 
D. MICHAEL DUGGAN APRIL 25, 1974 

Research such as that introduced by the above heading is normally hailed 
by the terminology of magnetic exchange. While commonly used as an indication 
that the exchange interactions are detected by techniques which measure mag- 
netic properties, this notation leads one astray from the source of these 
interactions, which is indeed electronic. The electron spin states of two 
metal ions perturb each other and lead to dimer states which have an energy 
distribution characteristic of the magnitude of this perturbation. This dist- 
ribution can be studied by various techniques such as magnetic susceptibility 
and epr, which measure the properties of electron spine. When refering to 
exchange coupling, the magnitude of the perturbation is parameterized by a 
value, J, and this arises from orbital interactions between the ions. The 
J value may be expressed in terms of coulomb and exchange integrals of the 
operator e /r. ., and of overlap integrals between the various orbitals of the 
metal ions and the bridging groups. 

It should be stated immediately that the detailed mechanism of exchange 
interactions is not understood, nor is it likely to be understood in the near 
future. Such a level of comprehension implying the ability to predict or 
calculate J values: it would clearly be necessary to possess an understanding 
of bonding and orbital properties much more complete than todays state-of-the- 
art, particularly when dealing with large systems. 

Why then would one want to undertake a study of magnetic exchange? First 
and primarily because the effects observed are intriguing; and the qualitative 
information that may potentially be gained about the bonding in a given system 
is unique to this method of study. S3Condly, one might consider that there are 
at least three general areas of vital chemical endeavor which owe their most 
interesting if not most important qualities to the presence of bridged metal.. 
clusters. Biochemical systems are known in many instances to involve more than 
one metal ion in the active site. In heterogeneous catalysis reactions of 
molecules on a, say, metal-oxide surface have been studied as a function of 
magnetic properties and the positive correlation between these phenomena and 
the observed reaction rates points out the possible importance of site-to-site 
electron exchange in the reaction sequence. Exchange coupling is certainly 
involved in the kinetics"of solution redox reactions. The role of bridging 
groups in the processes of inner-sphere and outer-sphere electron transfer is 
of great importance and it is clear that knowledge of exchange interactions 
and metal-metal exchange rates is fundamental to the understanding of this field. 

The generalized study of magnetic exchange for a series of metal ion 
systems may be conducted in several stages. The first of these, and often 
the most time consuming, is the preparation and determination of the structure 
of all materials. In view of the potentially large effects of small distance, 
and angle changes, X-ray techniques should be used; or at the least a careful 
spectroscopic comparison should be made between each unknown system and a 
closely related one whose structure is precisely known. Secondly the magnetic 
susceptibility and epr spectra of each material may be investigated, and all 
observed properties related to the electronic structure of the compounds, i.e. 
J values determined and any property of the data inconsistent with the quantum 
mechanical models should be investigated. The third step is to correlate J 
with the structural properties of each individual compound and most profitably 
with those for a series of related systems. This level is largely empirical 



in nature - allowing future predictions of structure based upon observed 
magnetic data. It is the fourth level that requires most often the molecular 
insight of the researcher. That is to say that here we seek an understanding 
of how exchange occurs and why it displays the observed trends. In the 
absence of detailed information about the molecular orbitals of a system, it 
is the researcher's qualitative judgement which arises here, and in general 
can lead to many possible interpretations of the same information. It is 
the collection of more and more empirical data that will cause a sorting of 
possible interpretations, and will in the future, with the simultaneous dev- 
elopement of quantitative techniques lead to the fifth level, scientific Nirvana. 

The particular study reported here involves the investigation of compounds 
with the following formula: 

[M2( amine )oXp]Y2, where 
M = Co(ll), Ni(ll), Cu(ll), 
amine - a tetradentate or tridentate ligand, examples: 

tren = 2, 2 '2" traminotriethylamine 

trien = triethylenetetramine 

dien = diethylenetriamine, 
X = bridging group = oxalate/2, squarate/2, Hj" , CM", OCN", SCN", SeCN", CI", and 
Y = counterion, most generally C10,~ or BPh/~. 

In order to supplement the battery of spectroscopic studies, three crystal 
structure determinations were carried out, those of [Ni 2 ( tren^CNCO^^BPh/ ^, 
[Cu 2 (tren)2(NC0) 2 ](BPh^)2 and [Cu 2 ( tren^CN^JCBPh^. Magnetic susceptibility 
and epr studies were carried out and all systems were characterized with respect 
to their exchange parameters. Level three results consist of the correlation 
of the observed J values with many properties, as listed below: 

1. For linearly bridged dimers, M-(ABC)2-M, J was correlated with the 
identity of the atoms and angular geometry of the bridge. 

2. Investigations were made into the dependence of exchange on the distortion 
of the single ion stereochemistry. 

3. J was measured as a function of extension of the bridge. 

4. The electronic configuration of the metal was studied with regard to the 
inherent exchange properties. 

5. Exchange through hydrogen bonds was detected and J was determined as a 
function of temperature. 

6. The effect of isotopic substitution of bridge atoms was investigated. 

Level four discussion of the results of this study has centered on the 
following considerations: 

1. The orientational and distance dependence of overlap between the metal 
and bridge orbitals. 

2. Symmetry considerations - i.e. the correspondence of the bridge orbitals 
used for metal bonding on one side of the bridge to those on the other. 

3. Sigma versus pi exchange mechanisms. 

<+. Overlap properties of MO's within the bridge. 

5. Relative energies of metal and bridge orbitals. 

6. Volume of the bridging MO's. 

In many cases fairly unambiguous interpretations of the observed effects 
have been made and in some instances it may be desirable to follow these up 
with single crystal epr spectroscopy to more fully characterize the single-ion 
orbital properties. 

It is hoped that in the future more concrete mechanistic decisions may 
be made on the basis of the data obtained through this study than are currently 
possible. 



poly(i-pyrazolyl)borate complexes and organometallics 

Robert Mink May 7, 1974 

Introduction 

A relatively new class of chelating agents consisting of boron 
and pyrazole derivatives, first developed by Trofimenko, is now be- 
coming the basis of an interesting research area in organometallic 
chemistry. These ligands can be complexed to transition metals, either 
in a bidentate or tridentate fashion. Depending on how they are com- 
plexed, their structures show different stereochemical nonrigidity as 
indicated by NMR spectra. Organometallic complexes involving the pyra- 
zolyl borate ligands are predominantly found with the metal in an 18 
electron configuration. 

Complexes 

The main classes of the boron pyrazole derivatives are the pyra- 
zaboles and the poly( l-pyrazolyl)borates. The pyrazaboles 1 * 2 are non- 
dissociable dimers of 1-boryl pyrazoles.* The six-membered ring BN 4 B 
is in a boat, 3 rather than planar conformation. NMR spectra show the 
equivalence of the 1,3,5*7 positions as well as the 2,6 positions. 
These compounds are very stable, being unaffected by air and water and 
can be stored for years without any apparent deterioration. When a 
BR 2 group in pyrazabole is substituted by a transition metal, the re- 
sulting class of compounds are the poly (l-pyrazolyl) borates. 4 These 
chelating ligands are uninegative and can be either bidentate or tri- 
dentate, depending on the number of pyrazolyl groups. 

The bidentate M[H 2 B(pz) 2 ] 2 complexes can form a square planar or 
tetrahedral configuration about the metal. 5 * 6 A planar array of nitro- 
gen atoms about the metal with the BN 4 M in a boat conformation means 
that the molecule can exist in two forms: a boat form with_both borons 
up and_a chair form with one up and one down. 7 The HB(pz.) 3 ~ and 
B(pz)4~ ligands in M[HB(pz)a] 2 and M[B(pz)4] 2 are tridentate and give 
rise to distorted octahedral complexes. > & 

Octahedral transition metal complexes with the configurations 
d 4 -d 7 can exist in two different electronic ground states, high spin 
and low spin, depending upon the strength of the ligand field. If, 
at a field strength A (octahedral splitting parameter) close to tt 
(mean spin pairing energy), the separation of the two states of dif- 
ferent multiplicity attains values within the thermally accessible 
range, equilibria between these states are expected to occur. This 
is the case for some Iron(ll) poly (l-pyrazolyl) borates. 9 * 10 Depending 
upon the substitution on the pyrazolyl rings, fully high spin, fully 
low spin, and complexes of intermediate spin can be produced. So- 
lution data and solid state data for a given complex at a given tem- 
perature will differ since, in the solid state, an additional component 
due to packing in the lattice increases the crystal field, thus 
favoring the low spin state in the solid. 1C Therefore, a complex at 
a given temperature can exhibit high spin- low spin equilibria in so- 
lution, but be diamagnetic in the solid state. A change in the crystal 
structure is expected in going from high spin to low spin. 



-59- 

The discovery of the poly(l-pyrazolyl) borate anions has opened 
up a new area of organometallic chemistry. The tris( 1-pyrazolyl)- 
borate ligand, RB(pz) 3 ~, forms numerous half sandwich complexes 11 
similar to the cyclopentadienide ion. However, the tris (l-pyrazolyl)- 
borate complexes appear to be more stable, both chemically and therm- 
ally, than their C 5 H 5 ~ analogs. One can change the electronic and 
steric effects without changing the symmetry of the molecule by substi- 
tution on the pyrazolyl rings. r ?rofimenko has synthesized bidentate 
(l-pyraeolyl)borato tt allyl dictrbonyl molybdenum complexes. 12 * 13 
These were assumed to be l6 eler^ron systems. There were two distinct 
classes, depending on whether the ligand contained alkyl or aryl groups 
on boron or in the 3,5 positions of the pyrazole ring. When the sub- 
stitution occurs in the pyrazole ring, the complexes are very stable 
and do not react with nucleophiles. However, when substitution occurs 
at the boron, the complexes resemble an electron-deficient structure 
by reacting with mcleophiles to achieve an 18 electron configuration. 
The difference; in reactivities for the two classes can be explained on 
steric grounds. X-ray crystallography has shown that B-H-Mo 14 and 
C-H-Mo 1 ^ 3 center, 2 electron bonds are involved. 

NMR 

The temperature dependent NMR spectra of poly(l-pyrazolyl)- 
borate transition metal complexes indicate different types of stereo- 
chemical nonrigidity, depending on whether the ligand is complexed in 
a tridentate, bidentate with a B-H-Metal interaction, or just simply 
a bidentate fashion. When the ligand is complexed in a tridentate 
fashion, the mechanism involves an internal rotation of the RB(pz) 3 ~ 
group around the boron-metal axis. 1 Attachment of bulky groups 
to the metal as well as substitution in the 3,5 positions of the 
pyrazolyl ring leads to a substantial increase in the rotational bar- 
rier. For the bidentate ligands complexed to the transition metal, 
the IR and NMR indicate the presence of conformational isomers. 19 * 20 
In solution, these conformers interconvert via ring flip mechanism. 
This mechanism also explains the fluxional character of some boron- 
substituted pyrazaboles. For bidentate ligands with a boron-hydrogen- 
metal interaction, IR spectra show that only one BN 4 M conformation 
exists in solution. The crystal structures show a pair of enantio- 
morphic molecules in the asymmetric unit. 14 In solution, the 
enantiomeric forms can interconvert, probably by a rotation of the 
pyrazolyl borate group around the metal. 13 * 21 

REFERENCES 

1. S. Trofimenko, J. Amer. Chem. Soc, 89, 3165 (1967). 

2. S. Trofimenko, J. Amer. Chem. Soc., H§, 4948 (1967). 

3. D. F. Rendle, A. Storr, and J. Trotter, J. C. S., Chem. Comm. , 

189 (1973). 

4. S. Trofimenko, J. Amer. Chem. Soc, 89, 3170 (1967). 



-60- 



5. L. J. Guggenberger, C. T. Prewitt, P. Meakin, S. Trofimenko, 
and J. P. Jesson, Inorg. Chem., 12, 

6. J. P. Jesson, S. Trofimenko, and D. R. Eaton, J. Amer. Chem. 
Soc, 89, 3148 (1967). 

7. S. Trofimenko, J. Amer. Chem. Soc., 9, (1967). 

8. M. R. Churchill, K. Gold, and C. E. Maw, Inorg. Chem., 9, 
1597 (1970). 

9. J. P. Jesson, S. Trofimenko, and D. R. Eaton, J. Amer. Chem. 
Soc, 89, 3158 (1967). 

10. J. P. Jesson, ,T. F. Weiher, and S. Trofimenko, J. Chem, Phys., 
48, 2058 (196b). 

11. S. Trofimenko, J. Amer. Chem. Soc, 21, 588 (1969). 

12. S. Trofimenko, J. Amer. Chem. Soc, D, 475 1 * 3). 

13. S. Trofimenko, Inorg. Chem., 9, 2493~Tl970). 

14. F. A. Cotton, M. Jeremic, and A. Shaver, Inorg. Chim. Acta, 6, 
543 (1972). 

15. F. A. Cotton, T. LaCour, A. G. Stanislowski, J. Amer. Chem. 
Soc, 9§, 754 (1974). 

16. S. Trofimenko, J. Amer. Chem. Soc, 91, 3183 (1969). 

17. P. Meakin, S. Trofimenko, and J. P. Jesson, J. Amer. Chem. 
Soc, 94, 5677 (1972). 

18. H. C. Clark and L. E. Manzer, J. Amer. Chem. Soc, 95 * 
3812 (1973). 

19. J. L. Calderon, F. A. Cotton, and A. Shaver, J. Organometal. 
Chem., 37, 127 (1972). 

20. J. L. Calderon, F. A. Cotton, and A. Shaver, J. Organometal. 
Chem., 38, 105 (1972). 

21. J. L. Calderon, F. A. Cotton, and A. Shaver, J. Organometal. 
Chem., 42, 4 19 (1972). 



-6l- 

HYDRIDE AND ALKYL DERIVATIVES OF NICKEL 

Alex N. Williamson May ?, 1974 

Recently, nickel hydrides and a-alkyls have received much at- 
tention in the area of homogeneous catalysis. Processes such as 
hydrosilation, hydroformylation, isomerization, and oligomerization 
have been postulated to involve nickel hydrides and alkyls as inter- 
mediates. In order to obtain a better understanding of the chemistry 
involved, much work has been done in the areas of synthesis and 
characterization of these compounds. 1 * 2 Today's seminar will deal 
mainly with these areas and, where possible, speculation will be made 
about the stability of these compounds . 

NICKEL HYDRIDES 

The first real evidence for nickel hydrides was reported by 
Green and co-workers in 1959. 3 A high field NMR signal was observed, 
but a compound could not be isolated, and it was ten years later be- 
fore a hydridonickel species was isolated and characterized. Today, 
there are over twenty- five examples of these species. They offer a 
wide variety of stereochemistries ranging from four coordinate trans 
planar 4 to five coordinate square pyramidal. 5 Most compounds are 
monomeric with terminal hydrides; however, there are examples of di- 
meric species with bridging hydrides. 1 * 6 

Methods for preparing hydridonickel complexes can be divided 
into three categories: (l) those prepared by beta elimination 7 of a- 
bonded nickel alkyls to give nickel hydrides and olefins; (2) those 
prepared by use of hydride transfer reagents 1 > 4 * 6 * 8 such as sodium 
borohydride and lithium borohydride, and (3) those prepared by oxi- 
dative addition of Bronsted acids to nickel (o) compounds. 5 ' 9 ' 10 Of 
all hydridonickel species studied, only the five coordinate hydrido- 
nickel phosphites are reported to show catalytic activity. These 
complexes are reported to catalyze both isomerization 11 and 
oligomerization 5 of olefins. 

NICKEL ALKYLS 

Alkyl nickel derivatives were originally thought to be too 
unstable for isolation. 12 Several of these compounds, however, do 
show remarkable stability to both air oxidation and hydrolysis. 13 
Compounds have been prepared by use of aluminum^* 14 * 15 * 18 lithium, 17 * 18 
and magnesium 13 * 19 alkyl reagents and also by oxidative addition, 10 
A wide variety of stereochemistries are observed for these compounds, 
as we have also noted for the hydridonickel species. The geometries 
range from three and four coordinate planar, 19 * 20 to five coordinate 
with 3,2 stereochemistry. 2 The five coordinate compounds, with the 
exception of the methyl derivative, decompose in the solid state al- 
most exclusively by beta elimination. 21 Planar mono- and dimethyl 
complexes react with phosphines, olefins and molecular nitrogen to 
form nickel (o) and nickel (l) compounds. 



-62- 

Structural characterizations of hydridonlckel and a-bonded 
alkyls have utilized primarily infrared and nuclear magnetic reso- 
nance spectroscopy. However, there are at least three examples of 
crystal structures. Isolation of these species seems to require 
the presence of tt acceptor type ligands, in that all compound iso- 
lated, contained either phosphines, phosphites or olefins. Common 
reactivity patterns and aerial decomposition products are yet to 
be established for these species. Since the field is yet in its 
infancy, much work remains to be done. 

REFERENCES 

1. K. Fisher, K. Jonas, P. Misbach, R. Stabba and G. Wilke, 
Angew. Chem. Intern. Ed., 12, 943 (1973). 

2. D. R. Fahey, Organometal. Chem. Rev. (A), 7, 245 (1972). 

3. M. L. H. Green, C. N. Street and G. Wilkinson, 
Z. Naturf orsch. , 14B, 738 (1959). 

4. M. L. H. Green, T. Saito and P. J. Tanfield, 
J. Chem. Soc. (A), 197 1 , 152. 

5. C A. Tolman, J. Amer. Chem. Soc, 92, 4217, 6777, 6785 (1970), 

6. K. Jonas and G. Wilke, Angew. Chem. Intern. Ed., 9, 312 (1970). 

7. S. C. Srivastava and M. Bigorgne, J. Organometal. Chem., 18, 
P30 (1969). — 

8. M. L. H. Green, H. Munakata and T. Saito, J. Chem. Soc. (A), 

1971* 46 9. 

9. K. Jonas and G. Wilke, Angew. Chem. Intern. Ed., 8, 519 (1969). 

10. R. A. Schunn, Inorg. Chem., 9, 394 (1970). 

11. C. A. Tolman, J. Amer. Chem. Soc, 94, 2994 (1972). 

12. J. Chatt and B. L. Shaw, J. Chem. Soc, I960, 1718. 

13. H. Yamazaki, Y. Matsumato, T. Nishide, S. Sumida and 
H. Hagihari, J. Organometal. Chem., 6, 86 (1966). 

14. T. Saito, Y. Uchida, A. Misno, A. Yamamoto, Merifuji and 
S. Ikeda, J. Amer. Chem. Soc, 88, 5198 (1966). 

15. G. Wilke and G. Herrmann, Angew. Chem. Intern. Ed., 5, 
581 (1966). 

16. P. W. Jolly, K. Jonas, C. Kruger and Y. H. Tsay, 
J. Organometal Chem., 33, 109 (1971). 

17. M. D. Rausch and F. E. Tribbets, Inorg. Chem., £. 512 (1970). 

18. M. L. H. Green and M. J. Smith, J. Chem. Soc. (A), 1971 , 639. 

19. B. Bogdonovic, H. Bonnemann and G. Wilke, Angew. Chem. 
Intern. Ed., 5, 582 (1966). 

20. B. L. Barnett and C. Kruger, J. Organometal. Chem., 42, 
169 (1972). — 

21. J. Thomson and M. C. Baird, Can. J. Chem., 48, 3443 (1970), 



04L& 



TABLE OF CONTENTS 
L 

fy£ INORGANIC SEMINAR ABSTRACTS 

I 197-'4-1975 

Summer Sessi on : Page 

STUDIES ON MIXED-VALENCE COMPOUNDS - 1 

William H. Morrison, Jr. 

STRUCTURE AND REACTIVITY OF SOME UNSATURATED 

ORGANOLITHIUM COMPOUNDS - 3 

Michael R. Walczak 

THE SYNTHESIS, STEREOCHEMISTRY, AND REACTIONS OP CO(lIl) 

COMPLEXES WITH AMINO ALCOHOL LIGANDS -- 4 

Martha S. Okaraoto 

Fall S e ssion ; 

THE DECOMPOSITION OF NICKEL(lIl) COMPLEXES OF SOME 

MACROCYCLIC TETRAMINE LIGANDS IN BASIC SOLVENTS - 5 

Michael T. Mocella 

SYNTHETIC APPLICATIONS OF METAL VAPORS - 6 

M. D'Aniello, Jr. 

ORGANOMETALLIC CHALCOGEN CLUSTERS - 9 

David J. Kitko 

THE CHARACTERIZATION OF THE 2p X-RAY PHCTOELECTRON SPECTRA 
OF FIRST ROW TRANSITION METAL ELEMENT COMPOUNDS - 
Gregory Allen Vernon 

MULTIPLE-QUANTUM ABSORPTION SPECTROSCOPY - 12 

Ronald G, Wollmann 

KINETICS AND MECHANISMS OF METAJ, CARBONYL REACTIONS - 15 

Jim Atwood 

POLYMER SUPPORTED GROUP VIII TRANSITION METAL 

IOGENEOUS CATALYSTS - It- 

Steve Richter 

CALCULATED TRENDS IN DINITROGEN BONDING '- 19 

John Gaul 

STUDIES ON THE FERRIC DITHIOCARBAMATE SPIN EQUILIBRIUM SYSTEM - 22 

Gretchen Hall 

AN ELECTRON SPIN RESONANCE STUDY OF SOME COBALT(ll) C( [ES - 23 

Ben Tovrog 



THE LIBRARY 7 

JUL 1 5 1975 






-2- 



THE EFFECTS OF METAL AND LIGAND VARIATION ON THE MASS 

SPECTROMETRY FRAGMENT'ATION OF TRANSITION METAL COMPLEXES - 24 

Jan Collard 

METAL COMPLEXES OF FOURTEEN-MEMBERED MACROCYCLIC LIGANDS 

CONTAINING TERTIARY AMINE DONORS - 25 

Frank Wagner 

INORGANIC APPLICATIONS OF RESONANCE RAMAN 26 

David S. Bieksza 

ELECTRON-NUCLEAR DOUBLE RESONANCE (ENDOR) OF TRANSITION 

METAL COMPLEXES 29 

Timothy R. Felt house 

MECHANISMS OF SUBSTITUTION REACTIONS OF METAL CARBONYL 

HYDRIDE COMPLEXES - 32 

Blaine H. Byers 



TODIES Q}\ MIXED-VALENCI C0MP0U1 



William H. Morrison, Jr. June 27 , 197^ 



Mixed-valence compounds are those which contain the same 
clement in formally two different oxidation states. ''Tie proper- 
ties of these compounds are rarely just tine sum of the properties 
of the two metal ions taken separately because of the inter-action 
between the two metal centers in the molecule. The properties of 
mixed-valence compounds will depend upon the amount of dereali- 
zation of the valence electrons, and thus on the extent of inter- 
action between the two moieties. 

Mixed -valence compounds have received a great deal of at- 
tention lately for several reasons: l) the close analogy between 
mixca-valence compounds and inner- and outer-sphere complexes formed 
during electron transfer reactions; < ) increased interest in under- 
standing biological mixed-valence* compounds, most notably the ferre- 
doxin systems; 3) interest in making high temperature superconduct- 
ing solids and good semiconducting materials; k) and finally one 
can learn a great deal about eletronic structure by studying mixed- 
valence compounds. 

Vfe have been interested in organoraatallic mixed-valence 

systems of the type 




rS)- x _y$j ^ 



Fe F' 

0>- *-<§) 

I II 

These compound were chosen because: l) they are readily adaj;table 
to structural change, i.e. one can modify * x 1 , or bridge one or 
both rings. By doing this, the effect of metal-metal distance and 
ring conjugation can be studied; 2) iron is known to have several 
easily accessible oxidation states, and ferrocene in particular 
is known to be easily oxidized; l) iron is the important biologi- 
cal atom; k) a lot is known about the electronic structure of 
ferrocene ana ferric enium. 

when biferrocene ( ., x-0) is oxidized to the raonooxidJ :ed 
form (n-l), it exhibits in its electronic spectrum a feature in the 
near ir not found in the parent molecule (n-0), or Ln the dioxi- 
di zed molecule (ns2). Sii.ii arly bifcrrocenyl >n< ..II, x*u), show:; 
a feature in its rnoaooxiui ed form, not* found in ti e parent ferro- 
cene or in the dioxidizeu form. These fea cures have been assigned 
to the IT (mixed-valence or intervalence transfer absorption; tran- 
sition in these molecules. A variety of physics] measurements 
(Mossbauer, esr, magnetism, r, and photo* lectron spectroscopy) 
have shown that biferriconiwa* das trap ied or localized valences, 
while biferricenylenium" r has average or del oca Lzoci valences. 



-2- 



Fitting of the physical properties tc a crystal field model 
gives an indication as to vhy this is the case. There appears to be 
a ground state change in going from biferricenium^ to blfcrricenyl- 
eniua -1 * . In bif erri cenyleni^m *" the electrons are much more delocal- 
ized due to either direct or super exchange . The dloxidized f'roras of 
these molecules strongly emphasize this, vith bif erricenium**^ bAM^ng 
a,. susceptibility curve similar tc- ferricenium, vhile biferricenyl- 
enium^ 2 is diaraagnetic . 

Gome (l.l) ferrorenophanes (II, x~CH-, CHCH3) have also been 
investigated. No band readably assignabl e to an IT transition was 
found in the raonooxidized ^orm of these compounds. The magnetism 
and esr properties nrf» similar to ferricenium systems, but the 
Mossbauer shows anomalies which ere at present not completely 
understood. The dioxiUzed form of the 1,12 dimethyl (l.l) fer- 
rocenox>nane (II, x*CHCH ) is diamugnetic, and has a Mtissbauer spec- 
trum similar to bif erricenylenium*'' . 



- o- 



STRUCTURE AND RFACTIVITY OF SOME UNSATURATED 
ORGANOLITiriUM COMPOUNDS 

Michael R. Walczak June 28, 197^ 

Organolithium compounds, such as n- butyl lithium, are useful 
reagents for many reactions, including anionic polymerization re- 
actions. Complexes containing an amine-solvated lithium cation 
coordinated to a tt delocalized carbanion form an especially important 
class of organolithium compounds in this chemistry. Based on nmr and 
x-ray structural studies of these unsaturated systems, a directed co- 
valent bonding model has been proposed to explain the relative orien- 
tation of the solvated lithium atom and the tt delocalized carbanion in 
the solid state. 1 Crystal structures of four examples of organo- 
lithium compounds- -bifluorenyl bis (lithium tetramethyl ethylenediamine) , 
stilbene bis (lithium tetramethyl ethylenediamine), stilbene bis- 
( lithium pentamethyldiethylenetriamine), and dilithioferrocene penta- 
methyl diethylenetriamine--will be presented, and some oxidation- 
reduction reactions of the delocalized carbanion systems will be 
discussed. 

Bifluorenyl Li^TMEDAp consists of two Li-TMEDA moieties associated 
with a bifluorenyl dianion. The geometry about each lithium atom is 
consistent with sp 2 hybridization, with one hybrid orbital directed 
toward the midpoint of the C9-C9' bond in the bifluorenyl carbanion. 
The remaining lithium 2p orbital bridges in a tt fashion between one 
ring carbon p orbital on each of the two fluorenyl rings in the bi- 
fluorenyl carbanion. Both structures containing the stilbene group 
contain base- solvated lithium atoms above and below the plane of the 
stilbene molecule. The stilbene molecule exhibits a trans configur- 
ation about the C7-C7' bond in each structure. Ferrocene Lip PMDTA 
exists as a dimer in the solid state, with two lithium atoms bridging 
between cyclopentadienyl rings from two ferrocene moieties. There are 
two types of lithium atoms in the dimer--a solvated lithium atom having 
all three PMDTA nitrogen atoms coordinated to it, and an unsolvated 
lithium atom bridging between ferrocenes. 

Organolithium compounds containing a tt delocalized carbanion are 
useful in preparing aromatic hydrocarbons 2 and coupled products. 
Molecular oxygen and transition metal halides and oxides convert the 
dianions to the neutral hydrocarbons, while the rnonoanio ns couple to 
give neutral dimeric products. 

References: 

1. G. Stucky, Advan. Chem. Ser. , 150 , S6 (1973). 

2. R. G. Harvey, L. Nazareno, and H. Cho, J. Amer. Chem. Soc, 

95, 2376 (1973). 



S, STZ: {, A] CI) COMPLEXES 

COHOL LI( 

Martha S. Okamoto July 2$, lyjk 

Although alcohol groups are not normally ce red to be good 
donors in metal o 5 , there is substantial evidence to indicate that 
atypical metal-oxyg f.eractions occur wh< .. alcohol group is in 
close proximity to a strong donor, such as an amine. In this work the 
Co(lll) complexes of l,3-diamino-2-propanol (2-tmGH) and 2,3-diamino- 
1-propanol (1-tmOH) were studied. 

2-TmOH was found to coordinate predominantly as a tridentate ligand. 
The most readily obtained complex was trans - [Co( 2-tmO) jCl. No evidence 
for the tris species was ever observed. The bis complex undergoes a 
series of reactions in acidic solution resulting in the isolation of 
complexes with different degrees of protonation. The stereochemistry 
of these complexes was determined primarily by C-13 nmr spectroscopy. 
1-TmOH, on the other hand, behaves as a simple diamine and the tris 
complex, [Co(l-tmOH) ]Br is readily obtained. 

Solutions of the complexes of tmOH ligands in which a halide is 
coordinated and the alcohol function is dangling underwent color changes 
characteristic of hydrolysis reactions. It was of interest to determine 
whether or not the dangling hydroxyl group participated in these reactions. 
Coordinated and dangling hydroxyl groups are readily distinguishable by 
C-13 n ,v :r. Therefore C-15 nmr studies of the base hydrolysis of cis- 
(Cc(en) (M CH CH OH)X] (X = Cl", Br"), [Co(TACM)(l-tmbH)Cl]Cl7TTACN = 
1,^,7-triazScyclononane), and [Co(TACN)(2-tmOH)X]X 2 (X = Cl", Br ) were 
undertaken. 



THESIS REPORT: THE DECOMPOSITION OP NICKEL(lIl) COMPLEXES 
OP SOME MACROCYCLIC TETRAMINE LEGAIJDS IN BASIC SOLVENTS 

Michael T. Mocella September 17, 197^ 

The nickel (III) complexes of the secondary amine macrocodes 
cyclam and Me a [l^]aneN4 were found to be unstable in basic solvents 






eye lam Me 6 [ 14 ] aneN* 

such as water and pyridine. However, the course of the decomposition 
was quite sensitive both to the particular ligand and to the solvent 
used. In all cases the formation of a nickel(Il) ligand radical species 
by deprotonation of the nickel (ill) complex followed by intramolecular 
ligand to metal electron transfer was an important initial step in the 
decomposition. Chemical and spectroscopic evidence in support of such 
an intermediate will be presented. The decomposition of this inter- 
mediate proceeded both with and without the formation of unsaturated 
macrocyclic complexes. The nature of these decomposition pathways has 
been inferred by a combination of chemical and spectroscopic methods. 

The nickel (ill) complex of the tertiary amine macrocycle tetra- 
methylcyclam decomposed by different routes in water and pyridine than 

tetramethylcyclam 

those found for the .secondary amine complexes. This was as expected, 
since intermediate formation as found for the secondary amine complexes 
is not possible in this case. 

The implications of the results of this work on literature reports 
of metal ion-assisted ligand oxidations, metal- substrate redox pro- 
cesses, and the problem of the identification of the site of oxidation 
in metal complex oxidations, will be considered. 



-6- 

SYNTHETIC APPLICATIONS OP METAL VAPORS 

M. D'Aniello, Jr. November 5, 197^ 

The reactions of high temperature species have been studied for 
many years. Until about five years ago, these studies were confined 
to main group elements such as carbon, 1 boron, 2 and silicon. 2 More 
recently, attention has been focused on the transition elements. 
This has resulted in the development of a promising new synthetic 
technique for the preparation of metal- containing compounds. 

The general technique involves the low pressure, low temperature 
co-condensation of transition metal and substrate vapors. Reaction 
occurs in the intimately mixed condensed phase. Such isolated metal 
atoms are thermodynaraically activated by a large fraction of the heat 
of vaporization of the metal, and can undergo reactions that would 
normally be endothermic with respect to the bulk metal. Likewise, con- 
siderable kinetic activation is also expected. Reactions which are, 
at best, surface phenomena involving the bulk metal become nearly homo- 
geneous in the condensed mixture, with the correspondingly low kinetic 
barriers of atom-molecule reactions. The low temperature of the mix- 
ture, however, will have a large effect on reaction rate; thus reaction 
may occur cnly after the mixture is warmed. Substrate to metal ratios 
are important, since reaction with substrate must occur before metal 
polymerization. Large substrate to metal ratios are generally em- 
ployed as this increases the probability that a given metal atom will 
encounter a substrate molecule rather than another metal atom. 

Considerable effort has gone into the design of systems in which 
such metal- substrate vapor reactions may be carried out. Several sys- 
tems have been designed. These differ in the method of vaporization 
of the metal and the way metal vapor is brought into contact with the 
substrate on a cold surface. Systems thus far developed employ re- 
sistively heated crucibles, 2 electron guns, 2 * 3 and even lasers 4 to pro- 
duce metal vapor. Both fixed 2 and rotating 3 * 5 vessels have been em- 
ployed to bring metal and substrate together. 

A variety of complexes have been prepared using the metal vapor 
technique. Chromium,® molybdenum, T and tungsten 7 have been co-condensed 
with various arenes, yielding the corresponding bis-arene complexes. 
Good yields can be obtained for the known bis-benzene and bis-toluene 
complexes. Several new complexes have been prepared involving arenes 
with electron- withdrawing substituents, e.g., Ph-P, Ph-OMe, Ph-NMe 2 . 
Rates of formation vary from l-2g./hr. to 0.1-0.2g./hr., depending on 
the ease of vaporization of the metal being used. 

Good results have also been obtained with olefins. The new com- 
plexes Fe( l,5-cyclooctadiene) 2 , 5 Ki(l,3-butadiene) 2 , 9 and M(l,3- 
butadiene) 3 , 12 M=Mo, W, have been prepared by the metal vapor technique. 
The molybdenum and tungsten complexes may very well open new fields of 
low-valent chemistry for these elements. Almost all the known low- 
valent chemistry of these elements require the presence of carbon monox- 
ide as a stabilizing ligand. It is notable that the 1,3-butadiene 
complexes are stable solids, and were prepared at the rate of 
0.2-0.3g./hr. 



Trifluorophosphine complexes are also readily prepared by this 
new technique. Conventional preparations of these complexes involve 
long high pressure or photochemical reactions. Using the metal vapor 
technique, 5-10g. quantities of the trifluorophosphine complexes of 
most of the first row metals can be prepared 1 " in a few hours. Simi- 
larly, new mixed PF 3 /arene, PF 3 /olefin, and PP 3 /nitrosyl complexes 
have been prepared with ease. 4 * 6 * B * 10 » U 

Metal vapors can also be used to accomplish substrate transfor- 
mations without necessarily isolating a metal complex. For example, 
a new procedure 13 for preparing diboron tetrachloride from boron tri- 
chloride involves the use of copper atoms as dechlorinating agents. 
Co- condensation of copper atoms and BCI3 results in good yields of 
B2CI4 in up to lOg./hr. quantities. Older methods require electrical 
discharges through BC1 3 , with rates of formation being less than Ig./hr. 

Hydrocarbon transformations have also been observed. Iron 11 * 14 " 
and nickel 9 atoms cause cyclohexadiene to disproportionate. Chromium 6 
and nickel 9 atoms will trimerize acetylenes to alkyl benzenes, and 
isomerize olefins. There are reports of ill- defined reactions between 
olefins and several other metals, among them aluminum, 15 magnesium, 16 
and platinum. 1T 

The variety of complexes prepared and uses employed thus far demon- 
strate that the metal vapor technique has considerable utility. Using 
this method, isolable quantities of compounds can be obtained, which 
by conventional techniques require rigorous conditions and considerable 
work-up time. The relatively short reaction times and, in particular, 
the cleanliness of reactions, recommend the technique for difficult 
syntheses and a basic research tool for preparation of new complexes 
unattainable by conventional routes. 

REFERENCES 

1. P. S. Skell, J. J. Havel, and M. J. McGlinchey, Ace. Chem. Res., 

6, 97 (1973). 
?. P. L. Timms, Adv. Inorg. Chem. Radiochem. , lh, 1?1 (197?). 
3. F. W. S. Benfield, M. L. H. Green, J. S. Ogden, and D. Young, 

J. C. S. Chem. Comm. , 1975 , 866. 
h. E. K. Von Gustorf, 0. Jaenicke, and 0. E. Polansky, Angew. Chem. 

internat. Edit., 11, 53? (197?). 

5. P. L. Timms and R. Mackenzie, J. C. S. Chem. Comm. , 1974 , 650. 

6. P. S. Skell, D. L. Williams-Smith, and M. J. McGlinchey, 

J. Amer. Chem. Soc, 95, 3337 (1973). 

7. M. P. Silvon, E. M. Van Dam, and P. S. Skell, ibid., 96, 

19^5 (1974). — 

8. R. Middleton, et. al., J. Chem. Soc. Dalt., 1973 , 1?0. 

9. P. S. Skell, et. al. , J. C. S. Chem. Comm., 197?, 1098. 

0. P. L. Timms, J. Chem. Soc. (A), 1970 , ?5?6. 

1. D. L. Williams-Smith, L. R. Wolf, and P. S. Skell, 
J. Amer. Chem. Soc., 94, 404? (197?). 

2. P. S. Skell, E. M. Van Dam, and M. P. Silvon, ibid., 96, 6?6 (1974). 



13. P. L. Timms, J. Chem. Soc, Bait., 197?, 830. 

14. P. L. Timms, J. C. S. Chem. Comm. , 19b9 , 1033. 

15. P. S. Skell and L. R. Wolf, J. AmerTcHem. Soc, 9^, 7919 (197?). 

16. P. S. Skell and J. E. Girard, ibid., 94, 5518 (197?). 
17.. P. S. Skell and J. J. Kavel, ibid., 9"37 6687 (1971). 



-9- 



ORGANOMETALLIC CHALCOGEN CLUSTERS 



David J. Kitko 



December 3, 1974 



In recent years, a large number of organometallic compounds 
have been prepared which contain two or more metal atoms in the 
same molecule. An important class of these compounds are the organo- 
metallic chalcogen clusters. 1 "" 19 These compounds contain a group 
VIA element as a bridging atom. A large number of these compounds 
have been prepared and characterized by Dahl and co-workers. 2 ~ 4 > 8 ~ 15 
They have also developed a qualitative model for explaining the metal- 
metal interactions through an LCAO-MO approach." 



2,3 



The trinuclear and tetranuclear chalcogen clusters of Fe, Co, 
and Ni are perhaps the most thoroughly studied of these compounds and 
will form the basis for this presentation. The trinuclear clusters 
prepared to date fall into two subclasses; those containing CO as the 
organic ligand and those containing a pentahapto-cyclopentadienyl 
group. The general molecular formulas are: 



M 3 (CO) 9 X 
M 2 M» (CO) 9 X 
M 3 cp 3 X 
M3CP3X2 



where M = Co; X = S, 1 ' 2 Se ; 



where M 
where M 

where M 

M 



Co; M 1 = Fe; X = S, 5 Se, 3 Te : 
Ni; X = N-C(CH 3 )3 6 ' 7 * 



Ni; X = 
Co; X = 



S, 8 CO 16 

36, T, 10 



M3CP3XY where M = Co; X = S; 10 Yz = CO 

X = O; 9 Y = CO 

All these trinuclear species contain a triangular array of metal atoms 
with the chalcogen acting as a triply bridging group above and/or 
below the plane of metal atoms. 

The tetranuclear species have the general molecular formulas: 



M 4 cp 4 X 4 



M4(N0) 4 X 4 



where M = Fe, 11 Co; 12 * 13 X = S 
M = Cr; X = 0, S 19 

where M - Fe; 14 X = S* 

M = Co; 14 X = N-C(CH 3 ) 3 * 



In these tetranuclear species, the chalcogen atoms are again triply 
bridging and the metal atoms are arranged in a tetrahedral array 
yielding a cubane-like cluster. Distortions from a purely tetrahedral 
arrangement of metal atoms have also been observed in these clusters. 11 



Not an organometallic chalcogen cluster. 



-10- 

The syntheses of these cluster compounds do not follow any 
simple trends. A variety of extended, high temperature, high 
pressure reactions have been used, and the compounds are often iso- 
lated in low yields. The starting materials are usually an easily 
obtainable organometallic species (Co 2 (C0) 8 > Fe(C0) 5 , (cpNiC0) 2 , etc.) 
and a chalcogen source (H 2 S, S, H 2 Se, or organic sulfur containing 
species). The clusters have reasonable solubility in organic solvents 
and are often separated from unreacted starting materials and unde- 
sirable by-products via column chromatography. An unusually large 
number of these compounds are air stable. Their characterization has 
involved the use of almost all available physical methods from the 
determination of their molecular weights via the molecular ion peak 
in their mass spectrum to the determination of their molecular struc- 
ture via X-ray crystallography. 

The extensive structural analysis of these complexes has spurred 
the development of the "metal cluster model" 2 * 3 for treating the 
metal-metal bonding interactions. This model proposed by Dahl and 
co-workers utilizes all the metal 4s, 4p, 3d orbitals to generate a 
set of metal symmetry orbitals. The available ligand orbitals are 
used to generate a set of ligand symmetry orbitals. Those orbitals 
with the correct symmetry and best overlap are combined to form the 
metal-ligand bonding and anti-bonding molecular orbitals. A perfect 
pairing approximation is assumed such that the metal-ligand inter- 
actions can be separated from the metal-metal interactions. The re- 
maining metal symmetry orbitals are combined to generate the metal- 
metal bonding and anti-bonding molecular orbitals. The relative energy 
level ordering is defined by overlap considerations and nodal charac- 
ter. The net metal-metal interactions are thus determined by the 
number of electrons which must be accommodated in these orbitals. Al- 
most the complete range of interactions from totally bonding clusters 9 
to those with no net metal-metal bonds 13 have been observed. 

This simplified treatment is not valid for a rigorous quanti- 
tative description of the electronic structure of these molecules. 
It nevertheless appears to explain the existing stereochemical data 
(i.e., metal-metal distances and geometrical distortions). It should 
be noted that complications arise in this method when strong rr- 
accepting ligands such as NO and CO are present, however, these can 
be accommodated. 14 * 15 

A number of these clusters possess interesting chemical and 
physical properties. Co 3 (C0) 9 S is a paramagnetic, air sensitive 
compound, and an ESR study of it in solution and doped into single 
crystals of the diamagnetic host FeCo 2 (C0)gS has shown that the un- 
paired electron is in a non-degenerate molecular orbital of a 2 
symmetry comprised primarily of an anti-bonding combination of d 
orbitals in the plane of the Co atoms. 2 Co 3 cp 3 S 2 is an air stable 
compound which exhibits a change in spin state from triplet to sing- 
let as the temperature is lowered to below 195°K. 17 It is readily 
oxidized by I 2 to yield a cationic cluster which is a simple paramagnet. 
The crystal structure of this cationic complex indicates that the 
cluster has undergone a distortion to C 2v geometry which is predicted 
by the metal cluster MO model and Jahn-Teller arguments. 10 The tetra- 
nuclear cluster Fe 4 cp 4 S 4 has a distorted D 2d geometry. This cluster 



exhibits stability in a series of oxidation states from 3 to -1 and 
characterization of the charged species is presently in progress. 18 

The organometallic chalcogen clusters are an interesting series 
of compounds which have provided valuable information concerning 
intramolecular metal-metal interactions. There are extensive efforts 
underway to synthesize and characterize new organometallic chalcogen 
complexes as well as other metal cluster compounds. These investi- 
gations will further illuminate these intramolecular interactions and 
perhaps help to determine their influence on the chemical reactivity 
of these compounds. 

REFERENCES 

1. L. Marko, G. Bor, and E. Klump, Chem. Ind. (London), lli91> (1961). 

2. C. E. Strousse and L. F. Dahl, Discuss. Faraday Soc, No. Ll7, 93 (1969). 

3. C. E. Strousse and L. F. Dahl, J. Araer. Chem. Soc, 93, 6032 (1971). 

h. D. L. Stevenson, C. K. Wei and L. F. Dahl, ibid., £3, 6027 (1971). 

£. S. A. Khattab, L. Marko, G. Bor, and B. Karko, J. Organomet. Chem., 
1, 373 (196U). 

6. S. Otsuka, A. Nakamura, and T. Yoshida, Inorg. Chem., 7, 261 (1963). 

7. S. Otsuka, A. Nakamura, and T. Yoshida, Justus Liebigs Ann. Chem., 719* 
$k (1968). 

8. H. Vahrenkamp, 7. A. Uchtmann, and L. F. Dahl, J. Amer. Chem. Soc, 90, 
3272 (1968). 

9. V. A. Uchtmann and L. F. Dahl, ibid., £1, 3763 (1969). 

10. P. D. Frisch and L. F. Dahl, ibid., 9h, 5082 (1972). 

11. C. H. Wei, G. R. "Wilkes, P. II. Treichel, and L. F. Dahl, Inorg. Chem., 
5, 900 (1966). 

12. G. L. Simon and L. F. Dahl, J. Amer. Chem. Soc, 95>, 216U (1973). 

13. V. A. Uchtmann and L. F. Dahl, ibid., 91, 3756 (1969). 

1U. ft. S. Gall, N. G. Connelly, and L. F. Dahl, ibid., 96, li017 (197k). 

15. R. S. Gall, G. T. W. Chu, and L. F. Dahl, ibid., 96, 1*019 (197)4). 

16. 3. 0. Fisher and C. Palm, Chem. Ber., 91, 172$ (19£8). 

17. M. Sorai, A. Kosaki, H. Suga, J. Seki, T. 'foohida, and S. Ctsuka, 
Bull. Chem. Soc Jap., hk, 236U (1971). 

18. J. A. Ferguson and T. J. Jfeyer, Chem. Comm. , 623, (1971). 

19. E. 0. Fisher, K. Ulm, and K. P. Fritz, Chem. Ber., £3, 2167 (I960). 

LIBRARY 
UNIVT7?S/Ty OF ILLINOIS 
AI WBANA- CHAMPAIGN 



THE CHARACTERIZATION OF THE 2p X-RAY PHOTOELECTRON SPECTRA 

OF FIRST ROW TRANSITION METAL ELEMENT COMPOUNDS 

Gregory Allen Vernon, Ph.D. 

Department of Chemistry 

University of Illinois at Urbana, Champaign, 1974 

The x-ray photoelectron spectra of the 2p shell of transition metal 
halides, pseudohalides , cyanides, and other complexes are studied. The 
data, together with literature data, are compiled to give an overall view 
of the phenomenon of satellite structure. 

The effects of varying the ligand, the central metal atom, and the 
charge on the central metal atom are analyzed. 

The data are explained on the basis of a charge transfer model 
where a ligand a(e ) electron is transferred into a metal 3d (e ) orbital. 

* 

This corresponds to a a(e ) -*■ a (e ) or 2e -*- 3e transition in molecular 

g g g g 

orbital terminology. 

The intensities of the satellites were found to be dependent upon 
the type of ligand and upon the energy spacing between the two e sets. 

A discussion of d systems is included with some spectral data from 
compounds of second and third row transition metal elements and lanthanides 



-12- 

MULTIPLE-QUANTUM ABSORPTION SPECTROSCOPY 

Ronald G. Wollmann February 11, 1975 

Introduction 

Multiple- quantum absorptions have been observed for many years 
in spectroscopic techniques spanning the energy range from high 
energy particle production 1 to electronic transitions 2 to electron 
paramagnetic resonance. 3 These processes involve the simultaneous 
absorption of two or more quanta. The transitions are completely 
allowed and do not arise from violations of single -quant urn absorption 
rules. Little experimental work has been directed toward the use of 
multiple -quant urn absorptions on an analytic scale comparable to that 
of single- quantum absorption spectroscopic methods. This is an un- 
fortunate situation since a multiple -quant urn absorption experiment 
is capable of yielding information about excited states of molecules 
which is unattainable from single-quantum absorption experiments. 4 
The use of multiple-quantum absorption has even been proposed as a 
technique for isotonic separation by selectively controlling isotopic 
chemical reactions. 

Theoretical 

Time-dependent perturbation theory 6 is the basis of theoretical 
treatments of the interaction of electromagnetic radiation with mat- 
ter. In 1931, Maria Gbppert- Mayer 7 presented the first quantum me- 
chanical treatment of the absorption or emission of two photons. In 
general, time-dependent perturbation theory will yield terms to j™ 
order which describe the absorption or emission of multiple quanta. 
Transition probabilities from these j"^ order terms can be obtained 
which describe the transition from the initial state n to the final 
state m. For example, the first and second order transition proba- 
bilities are 



P^(t)al(w)|<m|H'|n>| 2 g^) 

Pi!i l (t)al(a) 1 )l(uj 2 ) <m|H'|k'> <k'|fi'|n> g(a) 1 ,uu 2 ) 



n-*m 

where I (id) is the intensity of the incident radiation, g(uu) is a 
function of the frequency which contains the time-dependence of the 
perturbation and affects the lineshape of the transition, and H' is 
the spatial dependence of the perturbation. As can be seen from 

PiJft), the intensity of the two-quantum absorption depends on the 

square of the incident radiation intensity for uu^u^ and will be non- 
zero provided the matrix elements are non-zero. This gives rise to 
selection rules which are different than those for a single -quant urn 
absorption. For example, g<->g and u<-»u two-photon absorptions in 
centrosymmetric molecules together with polarization studies can be 
used to probe the excited states of molecules which can not be ob- 
served by single-photon absorption experiments. 8 



-13- 

Experimental Use 

Multiple -quant urn absorption experiments in the optical region 
of the spectrum have generally been used to verify predictions made 
by the band theory of metals, insulators, and semiconductors. 2 Only 
recently has this technique been extended to the study of organic 
molecules. 9 One-, ao two-, three-, and four- 11 photon absorptions have 
been observed in naphthalene and together with theoretical calcu- 
lations 12 can be used to assign the 1 B 1 g<— -Ag and 1 Ag^— 1 Ag transitions. 
No transition metal complex has been studied by two-photon absorption 
in the optica,^ region of the spectrum; but a two-photon absorption 
process has been indicated in the recently investigated dipyridinium 
pentachlorothallium(lll) molecule. 13 It was asserted that the process 
involved the initial absorption of two photons by the pyridinium ion 
followed by radiationless transitions to a Tl(lll) resonance state 
from which fluorescence was experimentally observed. 

Among the first multiple -quant urn absorptions observed in elec- 
tron paramagnetic resonance were those involving atoms such as 
oxygen 14 and chlorine. 15 More recently, multiple -quantum absorptions 
have been observed in sulfur 16 and fluorine. 17 Transition metal ions 
in octahedral symmetry have yielded easily interpretable multiple- 
quantum transitions. 18 * 19 Theoretical equations have been derived to 
calculate the resonant field of double -quantum absorptions and to 
calculate the zero-field splitting in the triplet states of organic 
molecules. 20 Although these equations were derived to explain the 
observed spectra of organic molecules, they have been successfully 
applied to transition metal complexes in triplet states. 3 > 21 > 22 The 
narrow linewidth of the double- quantum transition in a binuclear 
copper ( II ) complex enabled the accurate calculation of the zero-field 
splitting which otherwise would not have been possible with any degree 
of accuracy. 21 

Summary 

Multiple- quantum absorptions have been observed by many spectro- 
scopic techniques for many years. Chemists have just begun to recog- 
nize that useful information can be obtained from these processes. 
With the refinement of experimental techniques in the optical region 
of the spectrum, experiments with transition metal complexes will un- 
doubtedly become possible. Extension to states other than triplet 
states should also occur in the near future in electron paramagnetic 
resonance experiments. 



-14- 

References 

1. H. Terazawa, Rev. Mod. Phys., 45, 615 (1973). 

2. a) J. H. Worlock, in Laser Handbook, F. T. Arecchi, E. 0. Shulz- 

Dubois, Ed., North-Holland Publishing Co., Amsterdam, 1972. 
b) W. L. Peticolas, Ann. Rev. Phys. Chem. , 18, 233 (1967). 

3. G. A. Ward, B. K. Bower, M. Findlay, J. C. W. Chien, 

Inorg. Chem., 13, 6l4 (1974). 

4. W. M. McClain, Accounts Chem. Res., 7, 129 (1974). 

5. V. S. Letohkov, Science, 180, 451 (1973). 

6. G. Baym, Lectures on Quantum Mechanics, W. A. Benjamin, Inc., 

Reading, Mass., 1973. 

7. M. Gbppert-Mayer, Ann. Phys., 9, 273 (1931). 

8. A. M. Bonch-Bruevich, V. A. Khodovoi, Soviet Phys. Usp., 8, 

3 (1965). 

9. P. R. Monson, W. M. McClain, J. Chem. Phys., 56, 4817 (1972). 

10. J. B. Birks, L. G. Christophorou, R. H. Huebner, Nature, 217 , 

809 (1968). 

11. F. Pradere, J. Hans, M. Schott, Compt. Rend. Acad. Sci. Ser. A-B, 

265 , 372 (1966). 

12. H. E. Simmons, J. Chem. Phys., 40, 3554 (1964). 

13. G. P. Srivastava, S. C. Gupta, Optica Acta, 21, 43 (1974). 

14. V. ¥. Hughes, J. S. Geiger, Phys. Rev., 99, 1842 (1955). 

15. G. J. Wolga, Phys. Rev., 127 , 805 (1962). 

16. M. Jinguji, Y. Mori, I. Tanaka, Bull. Chem. Soc. Jap, 4j>, 1266 

(1972). 

17. C. A. McDowell, I. Tanaka, Chem. Phys. Lett., 26, 463 (1974). 

18. J. W. Orton, P. Auzins, J. E. Wertz, Phys. Rev. Lett., k, 128 

(I960). 

19. F. Chiarini. M. Martinelli, G. Ranieri, J. Chem. Phys., 4l 

1763 (1964). 

20. E. Wasserman, L. C. Snyder, ¥. A. Yager, J. Chem. Phys., 41, 

1763 (1964). 

21. J. Reedijk, D. Knetsch, B. Nieuwenhui jse, Inorg. Chim. Acta, 

5, 568 (1971). 

22. J. Reedijk, B. Nieuwenhui jse, Rec. Trav. Chim. Pays-Bas, 91* 

533 (1972). — 



-15- 



KINETICS AND MECHANISMS OF METAL CARBONYL REACTIONS 



Jim Atwood 



Final Seminar 



February 25, 1975 



One major area of study has been M(CO) 5 Br where M is Mn or Re 
and derivatives of these compounds. Mn(C0) 5 Br and Re(C0) 5 Br have 
both been studied with 13 C0 exchange and P(C 6 H 5 )3 substitution kin- 
etics. The reactions take place by a rate determining dissociation 
of radial CO, leading to a five- coordinate intermediate which is 
fluxional, and which rapidly picks up a ligand to reform the six- 
coordinate species. Cis Mn(C0)4P(C 6 H 5 )3Br has also been studied and 
shows a cis labilization by the P(C 6 H5)3. In cis Mn(CO) 4 P(C 6 H 5 ) 3 Br, 
CO dissociates from one of the two positions cis to both Br and 
P(CeH 5 ) 3 , at a rate which is five fold greater than the rate of dis- 
sociation from Mn(CO) 5 Br. The five-coordinate intermediate can be 
shown to be fluxional as well. These data suggest a previously un- 
noticed cis activation by weaker ir-bonding ligands. Coupled with data 
on the Mn(CO) 8 , we have the following sequence for dissociation from 
manganese carbonyls: 

Mn(C0) 4 P(C 6 H 5 )3Br > Mn(CO) 5 Br » Mn(CO)J 

Interpreting metal carbonyl reactions from the literature shows 
cis labilization to be a very general phenomenon and leads to the 
following ordering for cis labilization by ligands 

CO < H < P(C 6 H 5 )3 < I < Br < CI < py 
This is the reverse of the "trans effect" order. 

Cis labilization has many applications to reactivity of low valent 
metal complexes. In addition to the obvious applications to octahedral 
metal carbonyl reactions, there are possible applications to metal 
carbonyl cluster reactions and catalytic hydrogenations that proceed 
by oxidative-addition of H 2 to square planar complexes. 

The second major area of study has been the kinetics of substi- 
tution and exchange of Co 2 (CO) 8 . This molecule has obvious importance 
because of its catalytic properties, but kinetics studies have been 
complicated by the rapidity of its reactions. I have studied the ki- 
netics of exchange with 13 C0 at 0°C. and have found a significant rate 
dependence on the concentration of 13 C0. Observation of this unusual 
dependence on the 13 C0 concentration led to a study of the kinetics of 
Co 2 (CO) 8 reaction with other ligands such as P(C 6 H 5 )3. The rate of 
reaction of Co 2 (CO) 8 with P(C 6 H 5 )3 shows more than first order depen- 
dence on the concentration of Co 2 (CO) 8 and also a dependence on the 
concentration of P(C 6 H 5 ) 3 , indicating a complicated rate expression. 
Work is proceeding to determine the exact form of the rate expression. 
It appears unlikely, however, that these reactions are proceeding by 
a dissociative process as previously thought. 



-16- 

POLYMER SUPPORTED GROUP VIII TRANSITION METAL HOMOGENEOUS 

CATALYSTS 

Steve Richter February 27, 1975 

Catalysts are vitally important in today' s chemical process 
industries. The basic chemical technology used in fertilizer manu- 
facture, petroleum refining, petrochemical manufacture, or coal and 
coal-tar derivative processing involves processes that depend on 
catalyzed reactions. 

Historically, catalysts have been divided into two classes, 
namely, heterogeneous and homogeneous. In heterogeneous catalysis, 
a phase boundary exists between the catalyst and the reacting species 
(substrate). In homogeneous catalysis, all the species are in the 
same phase, most commonly the liquid phase. 

Heterogeneous catalysts are preferred in industrial operations, 
mainly because of the ease of separation of the catalyst from the 
reaction mixture. However, they have several drawbacks, because often 
they are not very specific, require high temperatures and pressures 
for operation, only have a small portion of their active sites readily 
accessible, and are difficult to design and improve. 2 

On the other hand, homogeneous catalysts are generally highly 
specific and highly active, react under comparatively mild conditions, 
and can have the steric and electronic environment of the active site 
widely varied. The great disadvantage of homogeneous catalysis is the 
difficulty with separation of the catalyst from the product, which 
often makes the process economically unfeasible. 2 Also, self- 
aggregation produces inactive species at high catalyst concentrations. 

Many of the problems of the two types of catalysts could be over- 
come if the best properties of each were combined. This combination, 
in effect, would be making a heterogeneous catalyst out of a homo- 
geneous one. In recent years, much effort has been devoted to making 
heterogeneous catalysts by linking homogeneous transition metal 
catalysts to solid supports. 3 * 4 

One type of solid support widely used for this purpose is an 
organic polymer. Most of these have been polystyrene derivatives, 
since polystyrene can readily be functionalized with a variety of re- 
active groups. In addition, since a large number of homogeneous 
transition metal catalysts contain phosphine ligands, the majority 
of these reactive groups have been phosphine derivatives. 5 

One system in which these polymer supports have been used in- 
volves Wilkinson's hydrogenation catalyst, RhCl(PPh 3 ) 3 . Grubbs and 
coworkers 6 have attached this homogeneous catalyst to a polystyrene 
support by phosphine links. The bound catalyst was easily separated 
from the reaction mixture by filtration and could be used many times 
with almost no loss in activity. A study of the rate of reduction 
versus size of olefin indicated that the polymer supported catalysts 
were selective for smaller molecules. Additional studies showed that 
this polymerized catalyst was regioselective in the hydrogenation of 
unsaturated steroids. 7 



-17- 

Strukul, etal, 8 found that polymer bound cationic Rh(l) complexes 
were selective hydrogenation catalysts of olefins and ketones. Chang- 
ing the substituents on the phosphorus atom of the polymeric ligand 
changed the catalytic behavior of the metal complex. 

Another system involves the hydroformylation catalyst Co 2 (C0)8 
and its phosphine substituted derivatives. Pittman and coworkers^ 
have synthesized a polymer supported catalyst similar to the phosphine 
substituted homogeneous Co 2 (CO) 6 (PPh 3 ) 2 . The bound complex had ap- 
proximately the same activity and selectivity as its homogeneous 
counterpart. 

It has been known for several years that Rhodium complexes, such 
as RhH(CO) (PPh 3 ) 3 , are more reactive and can be used under milder con- 
ditions for hydroformylation than the cobalt complexes. Pittman and co- 
workers 10 have also prepared the polymer supported analog of this 
homogeneous catalyst. The heterogenized catalyst had roughly the same 
activity and selectivity as the monomeric species, but the supported 
catalyst also retained its activity through many recycled batch 
operations. 

Care must be exercised, however, since the catalytic chemistry of 
polymer bound transition metal complexes can not always be inferred 
directly from that of the homogeneous analog. Side reactions can re- 
sult in the formation of surface functional groups on the polymer 
other than those intended. 11 Another problem with supported catalysts, 
one which they share with conventional heterogeneous catalysts, in- 
volves the very difficult task of characterization of the active sites. 
However, new techniques are being developed in surface science which 
might be of use in elucidating the structure and composition of the 
active sites in these catalysts. 12 

Overall, polymer supported homogeneous transition metal catalysts 
seem to combine the best advantages of both heterogeneous and homo- 
geneous catalysts. Some of these advantages are a) easy separation 
from the reaction mixture, possible regeneration, and reuse of the 
catalyst, b) less sensitivity of the catalyst to poisoning impurities, 
c) sometimes an exhibition of higher catalytic activity by the sup- 
ported catalyst, and d) great versatility due to the many possible 
modifications which can be made on the polymer support. 

References 

1. 0. F. Joklik, Chem. Eng . , 80, 49 (1973). 

2. K. G. Allum, R. D. Hancock, S. McKenzie, and R. C. Pitkethly, 
Catalysis , Vol. 1 (Proceedings of the 5th International Congress 
on Catalysis, Palm Beach, Florida, August, 1972) (J. W. Hightower, 
ed.), North Holland, New York, 1973, p. 30-477. 

3. J. C. Bailar, Jr., Cat. Rev.-Sci. Eng . , 10 , 17 (1974). 

4. E. Cernia and M. Coraziani, J. App. Poly. Sci ., 18, 2725 (1974). 

5. G. J. K. Acres, A. J. Bird, and P. J. Davidson, cHem. Engr. 
(London) , 285 , 145 (1974). 

6. R. H. Grubbs, L. C. Kroll, and E. M. Sweet, J. M acromol. Sci. -Chem., 
A7, 1047 (1973). 



-18- 



7. R. H. Grubbs and S. Phisanbut, in press. 

8. G. Strukul, M. Bonivento, M. Graziani, E. Cernia, and 
N. Palladino, Inorg. Chim. Acta , 12, 15 (1975). 

9. G. 0. Evans, C. U. Plttman, Jr., R. McMillan, R. T. Beach, and 
R. Jones, J. Orgmet. Chem . , Gj , 295 (1974). 

10. C. U. Pittman, Jr., and R. M. Hanes, Ann. N. Y. Acad. Sci. , 
in press. 

11. T. 0. Mitchell and D. D. Whitehurst, Presented at the Third 
North American Conference of the Catalysis Society, 

San Francisco, February, 197^. To be published in 
J. Amer. Chem. Soc . 

12. J. T. Yates, Jr., Ch. Eng. News ., 52, 19 (197*0. 



-19- 



CALCULATED TRENDS IN DINITROGEN BONDING 



John Gaul 



March 6, 1975 



Introduction 

Since the discovery of Ru(NH 3 ) 5 N2 2+ by Allen 1 in 19&5* tremen- 
dous efforts have been made to isolate and characterize new compounds 
containing N 2 . To date, there exists almost 200 such compounds en- 
compassing 18 transition metals. Their oxidation states range from 
zero 2 * 3 ' 4 to five 5 and can have coordination numbers in solution from 
three 6 to the more common 7 * 8 four and six. In addition, many bis-N 2 
and N 2 bridged systems exist. Formation of these bridged systems from 
mononuclear N 2 compounds indicates that the reactivity of the N 2 is ef- 
fected little by coordination. 7 ' 8 The frequency (\w) is usually 
lowered to -^-^1600-1700 cm" 1 (compared to free N 2 ~~ 2331 cm" 1 ), indi- 
cating a reduced bond order 9 and perhaps "activation" of the N 2 . 

Structure of the N 2 Moiety 

Almost all N 2 compounds known today consist of linear fragments 
in agreement with m.o. calculations of stability. 10 There are a few 
notable exceptions. An x-ray structure on a Ni dimer, 
[(C 6 H5Li) 4 (C6H5)Li 2 Ni 2 N 2 (Et 2 0) ] 2 , indicates the N=N bond-axis is at 
right angles to the Ni-Ni bond. 11 Bercaw reports dinitrogen tt bonded 
to a titanocene. 12 Most other edge-on bonded N 2 complexes have been 
found by matrix isolation techniques. 

Preparations and Reactions of Coordinated N 2 

Most dinitrogen compounds can be prepared in similiar fashions. 
Usually, the metal as an acetylacetonate is reduced (aluminum alkyl, 
Grignard reagent, Zn/Hg) in the presence of NH 3 or PR 3 and N 2 to yield 
the dinitrogen compound. 6 ' 7 

Red. Agt. 
M(acac) 2 + ligand ■■> M(N 2 )L 3 

N 2 

The ligand may be NH 3 , PR 3 , H 2 0, X", CO, H", 3° arsine or a 3° stibene. 

The reactions of coordinated N 2 for the most part consist of 
substitutions of the N 2 by CO, H 2 or some other ligand. There is, 
however, a growing number of N 2 - compounds that react to yield stable 
N-H and N-C bonds and these are generating considerable interest. 9 * 1 ' 4 ' 15 ' ie 
Some examples are 



(0 3 P)HRFeN 2 FeR(P0 3 ) 2 
W(N 2 ) 2 (DPPE) 2 + R0C1 
M.O. Calculations 



HC1 
-> complex + N 2 H 4 9 



■ > 



[(DPPE) 2 C1 2 W(NHNC0R) ] 



14 



Hopes of understanding why some dinitrogen compounds seem to acti- 
vate N 2 whereas others leave the N 2 little changed from free N 2 has led 
to theoretical calculations of varying degrees of sophistication. In- 
sight into dinitrogen bonding is obtained from stability predictions 



-20- 

from simple Huckel and calculations of bond order and charge distri- 
bution from CNDO and ab initio SCF methods. 

Though a simple Huckel calculation is of little value in a quanti- 
tative sense, when applied to dinitrogen compounds provides insight 
for predicting which metal and its appropriate oxidation state should 
result in greater stability. 13 They were also able to calculate curves 
for electron charge and bond order as a function of 6 (a term relating 
the energies of the tt orbitals of N to the d orbitals of the metal). 

These results are in accord with a subsequent calculation 1 " 7 of 
the same quantities by a different method. However, the latter calcu- 
lations are much more quantitative and broader in scope. Charges on 
the metal and each nitrogen were calculated along with degrees of a 
to drr-prr* bonding for metals in the 1st and 2nd transition series. 
Some very definite trends were observed. Calculated stabilities of 
the M-N 2 moieties were plotted for the first two transition series in 
various oxidation states. Trends in bond orders were also observed. 

I.R. and ESCA 



Experimental verification of the m.o. calculations for Fe, Ru, 
Os, Ir and Re was attempted with ESCA and IR. These metals were se- 
lected for a number of reasons: 

(l) they have been studied extensively 
{?-) Fe, Ru, Os are a triad and thus might show trends 
(3) Re, Os, Ir are a 5d sequence and it might be learned if 
the trends hold true for these metals. 

Charge separation data was obtained using correlation plots of 
Nls binding energies vs. calculated charges for nitrogens in various 
formal oxidation states. 18 * 19 Correlation diagrams corresponding to 
three calculation methods were obtained. Results with Re, Ir, and Fe 
seem to correspond to predicted trends whereas those for Ru and Os are 
inconclusive. Charge separation data calculated by another method 21 
for Ru and Os does seem to correlate. Metal core binding energies were 
measured for all five metals and trends in the magnitude of the 
positive charge tabulated. 

From the ESCA studies, it is seen that the N=N bond is quite 
polar 20 in N 2 complexes. The authors conclude that the charges in 
the nitrogens are opposite to the trends predicted by m.o. calculations, 

IR intensity values are interpreted in terms of the relative 
amounts of dn-pn* bonding to a-bonding in these compounds. The in- 
tensity to frequency 21 * 23 * 24 plots indicate a great deal of similarity 
between the Fe 20 and Ru 25 complex while the predicted trend is seen 
for Ru and Os and for Re, Os and Ir. These IR intensities were of 
value in providing a check on the method of charge assignment noted 
above. 

Thus, from experimental data (IR and ESCA), overall trends of 
charge separation, bond order and relative importance of dn-pir* bonding 
both across a series and doxvn a triad seem to agree with calculated 
trends. The convention of assigning charges to the two nitrogens is 
seen to be disputed with conclusions arrived at experimentally. 



-21- 

References 

1. A. D. Allen, C. V. Senoff, Chem. Comm ., 621 (1965). 

2. J. Chatt, R. H. Crabtree, J. Chem. Soc. Dalt. Trans . (1973), 11 , 
1167. 

3. G. A. Ogin, A. VanderVoet, Can. J. Chem ., 51* 3332 (1973). 

4. D. J. Darensbourg, Inorg. Chem ., 13, 1532 "(197*0. 

5. D. J. Darensbourg, Inorg. Chem. Acta , 6 (4), 527 (1972). 

6. D. Negoiu. C. Parlog, D. Sanderlescu, Rev. Roum. Chem ., 19* 
387 (1974). 

7. A. D. Allen, Adv. in Chem. Series , #100, 79 (1971). 

8. A. D. Allen, R. 0. Harris, B. R. Loescher, J. R. Stevens, 
R. W. Whiteley, Chem. Rev ., 73 (l), 11 (1973). 

9. Y. G. Borodko, M. 0. Broitman, L. M. Kachapina, A. E. Shilov, 
L. Y. Ukin, Chem. Comm ., II85 (1971). 

10. Y. A. Kruglyak, K. B. Yatsimirskii, Teor. i E'kperim. Khim . , 5* 
308 (1969). 

11. K. Jonas, Angew. Chem . (1973), 85, 1050; C. Kreuger, Angew Chem . 
. (1973), 85, 1051. 

12. J. E. Bercaw, JACS , 96 (2), 6l2 (1974). 

13. Scientia Sinica , 17 J2) , 193 (1974). 

14. J. Chatt, G. A. Heath, G. J. Leigh, J.C.S. Chem. Comm. , 444 (1972) 

15. J. Chatt, G. A. Heath, R. L. Richards, J.C.S. Chem. Comm ., 
1010 (1972). 

16. J. E. Bercaw, JACS , 96, 5087 (1974). 

17. S. M. Vinogradovo, Y. G. Borodko, Russ. J. Phys. Chem ., 47 (4), 
449 (1973). 

18. D. N. Hendrickson, J. M. Hollander, W. L. Jolly, Inorg. Chem ., 
8 (12), 2642 (1969). 

19. Borge Polkesson, Acta Chem. Scand . , 27, 287 (1973). 

20. Borge Folkesson, Acta Chem. Scand . , ^7 (4), l44l (1973). 

21. Borge Folkesson, Acta Chem. Scand . , 27 , 4008 (1972). 

22. Borge Folkesson, Acta Chem. Scand . , 27, 276 (1973). 

23. A. D. Allen, J. R. Steven, Can. J. Chem ., 50, 3093 (1972. 

24. Y. G. Borodko, et al. , Russ. J. Phys. Chem ., 44 (5), 643 (1970). 



-22- 

STUDIES ON THE FERRIC DITHIO CARBAMATE SPIN EQUILIBRIUM SYSTEM 

Gretchen Hall Final Seminar March 11, 1975 

The Fe(lll) dithiocarbamates have been extensively studied, 1 
however, the spin equilibrium system is still not well characterized. 
Iron-57 M^ssbauer studies have set a lower limit on the spin flipping 
rate of 10 7 sec" 1 , and this has led to the suggestion that these are 
mixed spin instead of spin equilibrium systems. 2 Our work has concen- 
trated on establishing a qualitative estimate of the spin flipping rate 
and on determining the details of the electronic structure of the 
ferric dithiocarbamates. 

Magnetic susceptibility data have been collected from 296-4. 2°K 
for ten Fe(lll) dithiocarbamates and for two Ru(lll) dithiocarbamates. 
The ferric dithiocarbamates exhibit a Boltzmann population distribution 
over the three Kramers doublets from the low spin ( 2 T 2 ) ground state 
and over the three Kramers doublets from the high spin (®Ai) excited 
state, however, the susceptibility data can not be fit with a simple 
2 T 2 - 6 A! equilibrium model. It is necessary to include a vibrational 
partition coefficient 3 (or a mathematically equivalent temperature de- 
pendent 2 T 2 - 6 A! energy separation) to fit the data. 

Infrared measurements from ambient temperature to 30°K on these 
systems have been used to l) determine the validity of using the vi- 
brational partition coefficient, and 2) set an upper limit on the spin 
flipping rate. The ir data for structurally analogous cobalt, ru- 
thenium, and manganese dithiocarbamates have been used to help in the 
understanding of the ir spectra of the iron systems. 

Epr studies from 85-4. 2°K have been made on these systems. 
Measurements on the completely high spin ferric dithiocarbamates and 
the low spin ruthenium dithiocarbamates have been used to interpret 
the epr work. 

The ir and epr results set an upper limit of 10 10 sec -1 on the 
spin flipping rate. The ir data also shows that the previous analysis 
of the susceptibility data based on either a vibrational partition co- 
efficient or a temperature dependent energy separation is incorrect. 
One possible alternative is a model involving extensive vibronic inter- 
action in the 2 T 2 levels. 

1) R. L. Martin and A. H. White, Trans. Metal Chem . , edited by 

R. L. Carlin, 4, 113 (1968). 

2) P. B. Merrithew and P. G. Rasmussen, Inorg. Chem ., 11 , 325 (1972). 

3) A. H. Ewald, R. L. Martin, E. Sinn and A. H. White, Inorg. Chem . , 

8, 1837 (1969). 



-23- 

AN ELECTRON SPIN RESONANCE STUDY OF SOME COBALT ( II ) COMPLEXES 
Ben Tovrog Final Seminar March 18, 1975 

Recent studies involving planar cobalt ( II ) complexes have revealed 
interesting reactivities with regards to adduct formation. Extensive 
studies on porphyrin and Schiff base systems have shown that in the 
presence of a coordinating Lewis base, a molecule of dioxygen will co- 
ordinate forming a six- coordinate complex. Speculations concerning the 
influence of the axial base and in-plane ligand set in the subsequent 
reaction with dioxygen have been made. Esr studies, however, which can 
be very useful in determining the electronic structure of these five- 
coordinate base adducts have been limited, likely due to the weak 
acidity of the cobalt center. 

In addition to the specific problem state above, the general area 
of the influence adduct formation has on the esr of transition metal 
complexes is of interest. Consideration of experimental limitations, 
relaxation effect's and known chemistry cause one to conclude that such 
studies can be effectively conducted only on first row S=l/2 low sym- 
metry systems. It is then understandable that much work has been done 
in the area of copper(ll) complexes, though many studies in this area 
have produced confusing and conflicting results. For these reasons, we 
have chosen to study BF 2 capped bis-diphenylglyoximecobalt(ll) , due to 
the fact that, in contrast to previously studied cobalt (II), this com- 
plex readily forms both 1:1 and 2:1 adducts with a wide variety of nitro- 
gen and oxygen donors. Our interest in this system was further height- 
ened by the fact that solutions of Co(DPGB) 2 with stoichiometric amounts 
of added base readily form dioxygen adducts at low temperature. The 
coordinated dioxygen, however, can be displaced with addition of excess 
base, forming the 2:1 base adduct. The BF 2 capped rather than the pro- 
ton bridged ligand was chosen for study due to the decreased basicity 
of the glyoxime oxygens which allowed study of adducts formed from weak 
donors, due to the decreased tendency for dimer formation. 

The esr studies have been conducted both at liquid nitrogen tempera- 
ture in glassy media and at room temperature in solution. All adducts 
are low-spin and characterized by an orbitally non- degenerate ground 
state, (cL 2 ) 1 . A crystal field treatment of the anisotropic spin- 
Hamiltonian g and A values has been used, with spin- orbit effects from 
excited states included as a perturbation. Ligand hyperfine have been 
similarly analyzed. 

Comparing our data with that for other systems, it is noted that 
the cobalt center in Co(DPGB) 2 has a larger <l/r 3 > (higher positive 
charge). This result is used to explain the interesting finding of in- 
creased nitrogen and decreased phosphorus hyperfine coupling when com- 
pared to Co(salen) adducts of similar bases. 

The wide variety of 2:1 adducts of both nitrogen and oxygen donors 
has revealed wide variations in the P values (Pa<l/r 3 >) characterizing 
the complex, indicating that variations in axial donors can signifi- 
cantly alter the electronic structure. Correlations between these P 
values, with E„ and C^ numbers characterizing the base, exist and pro- 
vide insight as to factors influencing charge distribution in these 
adducts. It has been found that the P value is very dependent on E„ of 
the axial base. C R has obvious importance because for bases with 
similar Eg, P decreases in the order donor > N donor > P,S donor. 



-24- 

THE EFFECTS OF METAL AND LIGAND VARIATION ON THE MASS SPECTROMETRY 
FRAGMENTATION OF TRANSITION METAL COMPLEXES 

Jan Collard Final Seminar March 25, 1975 

Although mass spectrometry has been in existence for quite some 
time, it is only beginning to be exploited by transition metal chemists. 
It has been found that the fragmentation pattern of an organic ligand 
changes upon complexation with a metal ion. 1 This is not surprising in 
light of the theory of mass spectrometry which predicts that a bombard- 
ing electron in the ion source interacts with a molecule causing it to 
emit an electron from the highest occupied molecular orbital (HOMO). 2 
Molecular orbital theory has shown that, in general, the HOMO of a 
transition metal complex is composed essentially of metal d orbital 
character. 3 Thus, the electron that will be emitted has a large per- 
centage of metal d character in the metal complex. 

The object of this particular study was to investigate the influ- 
ence of the metal d electronic configuration on the fragmentation of a 
transition metal complex in the mass spectrometer. To do this, a series 
of complexes with very similar ligands and varying metal d electronic 
configuration was needed. Such a series is the iron(lll) dithiocarba- 
mates of general formula Fe(S 2 CNR 2 )3. Recently, 4 the electronic struc- 
ture of these complexes was discussed, and it was shown that the elec- 
tronic state for these complexes varies with a change in the alkyl 
substituent R. As a control group, the cobalt and ruthenium analogs of 
the dithiocarbamates were also studied. The electronic configuration 
of these compounds remains constant throughout the series. 

The fragmentation patterns of these complexes are interesting and 
unique and the presence of the metal ion in the compound allows one to 
see types of fragmentation not observed for organic compounds. Such 
processes can be explained utilizing a valence change concept 5 which 
invokes the variability of valence state found for transition metals. 

As the amount of low spin character in these complexes increases, 
there is expected to be a consequent increase in the strength of the 
metal- ligand interaction. This increase in interaction should be re- 
flected in the mass spectra by an increase in the number and intensity 
of ions containing the intact metal- sulfur bond. However, there are 
many competing fragmentation pathways possible, making a linear corre- 
lation with metal- sulfur bond strength impossible. 

References 

1. M. J. Lacey, C. G. MacDonald, J. S. Shannon, Org. Mass Spec , I, 

115 (1968). 

2. R. A. W. Johnstone, Mass Spectrometry for Organic Chemists , 

Cambridge University Press, London, 1972. 

3. C. J. Ballhausen, H. B. Gray, Molecular Orbital Theory , 

W. A. Benjamin, Inc., New YorkJ 1965. 

4. G. Hall, Final Seminar, March 11, 1975. 

5. M. J. Lacey, J. S. Shannon, Org. Mass Spec . , 6, 931 (1972). 



-25- 

METAL COMPLEXES OF FOURTEEN -MEMBERED MACROCYCLIC LIGANDS CONTAINING TERTIARY 

AMINE DONORS 

Frank Wagner Final Seminar April 15, 1975 

A wide variety of macrocyclic ligands that contain some combinations of 
nitrogen, oxygen and sulfur donors have been synthesized and studied during 
the past fifteen years. 1 The majority of the macrocycles studied are fourteen- 
membered ring systems with nitrogen donors. Very little attention, however, 
has been given to macrocyclic ligands that contain tertiary amine donors. ^>3 

Initially, a fully alkylated fourteen-membered macrocyclic complex of 
Ni(III) was required for redox studies.^ Reaction of cyclam with formic acid 
and formaldehyde (Eschweiler-Clarke reaction) afforded the tetramethylated 
derivative (TMC) in good yields. Treatment of Ni 2+ with TMC in ethanol led 
to the isolation of the red, crystalline complex, NiTMC(C10],)2. The complex 
exhibited an unusual lability in dilute mineral acids, rapidly decomposing 
to aqueous Ni 2+ and protonated ligand. In addition, the ligand was completely 
transferred from nickel to either Cu 2+ or Ag in aqueous solution. Such a 
lack of kinetic stability could not be explained in terms of the donor properties 
of the tertiary amine but must result from the stereochemistry of the ligand 
about the metal center. Of the five possible sets of nitrogen configurations, 
the physical and chemical properties of NiTMC 2 + and analogous Zn 2+ complexes 
suggested that all four N-methyl groups were on the same side of the molecular 
plane .5 This is not the thermodynamically favored form (Nicyclam 2+ structure) 
but is a kinetically controlled product. A crystal structure determination 
on Ni(TMC)N3 + confirmed the assignment of stereochemistry. 

Attempts to synthesize the thermodynamically stable form led to develop- 
ment of procedures for alkylating deprotonated species. Treatment of Nicyclairr + 
with KOH in DMSO followed by addition of methyl iodide yielded a tetramethylated 
product quite different in properties from NiTMC 2 "*". A crystal structure deter- 
mination of a mono-azido bridged dimer indicated that the ligand possessed the 
cyclam-like stereochemistry which is the thermodynamically most stable form. 7 
The deprotonation reactions were applied to other macrocyclic systems utilizing 
sodium methylsulfinylmethide in DMSO as the base -solvent system. Alkylation 
reactions were restricted to alkyl halides without beta -hydrogens since 
elimination competed with substitution. Most alkylations were stereospecific, 
but in some cases rapid proton exchange among nitrogen sites in the deprotonated 
species dictated the distribution of products. Attempts to unequivocally 
determine the stereochemistry of a variety of alkylated products from the 
aforementioned reactions led to development of a new "template" synthesis 
of fourteen-membered macrocycles. 

References 

1. James J. Christenen, Delbert J. Ea tough and Reed M. Izatt, Chem . Rev ., 7k* 

351 (197U). 

2. G. A. Kalligeros and E. L. Blinn, Inorg . Chem ., IT, 11h$ (1972). 
3e R. Buxtorf, W. Steinman and T. A. Kaden, Chimia, ,28, 19 (197U). 
U. E. K. Barefield and M. T. Mocella, to be published. 

5. E. Kent Barefield and F. Wagner, Inorg . Chem ., 21+35 (1973). 

6. M. J. D'Aniello, Jr., M. T. Mocella, F. Wagner, E. Kent Barefield and 

Iain C. Paul, J. Amer . Chem . Soc , 21, 192 (1975). 

7. F. Wagner, M. T7 Mocella, M. J. D'Aniello, Jr., A. H. J. Wang, E. Kent 

Barefield, J. Amer . Chem . Soc, 96, 2625 (197U). 



-26- 

INORGANIC APPLICATIONS OF RESONANCE RAMAN 

David S. Bieksza April 24, 1975 

Introduction 

The Raman effect is essentially a light-scattering phenomenon. 
A portion of the incident radiation is scattered by a molecule at 
altered frequency according to the molecule's characteristic energy 
levels. Depending on the position of these energy levels, the fre- 
quency shifts in the scattered radiation correspond to rotational, 
vibrational, or electronic transitions. 1 The resonance Raman effect 
involves the relationship of the incident radiation to the frequencies 
of electronic transitions in the molecule. Two separate cases may be 
distinguished. The pre-resonance effect occurs when the incident fre- 
quency approaches a transition frequency. In this case, the intensi- 
ties of certain peaks in the Raman spectrum increase compared to the 
rest. The rigorous resonance case, on the other hand, occurs when the 
incident frequency falls within the band envelope of the transition. 
The enhancement of intensity is so great that overtones and occasionally 
combination bands appear. 2 

Theory 

The intensity of scattered radiation for a freely rotating mole- 
cule at frequency v £ over the entire solid angle is proportional to 
the intensity of the incident radiation, the fourth power of the 
scattered frequency (hence the "v 4 law") and the sum of the squares 
of the elements of the polarizability tensor. 3 However, the function- 
al dependence of the elements of the polarizability tensor has been 
derived employing two different approaches. Dispersion theory, modi- 
fied for resonance conditions, yields an equation which relates each 
element of the tensor to a sum of virtual excited electronic states, 
taken singly, and all the vibrational sub- levels of each state. 
Another derivation, through time-dependent perturbation theory, 4 yields 
an equation relating each element of the polarizability tensor to a 
sum of virtual excited electronic states, taken pairwise. Super- 
ficially, both equations explain the enhancement of intensity in the 
same manner: As the incident frequency approaches the frequency of 
an electronic transition, the denominators of both expressions de- 
crease, thus intensity increases. A damping constant, related to the 
half-width of vibrational fine structure in an electronic absorption 
band, prevents infinite intensity in the rigorous resonance case. The 
point should be made here that both summations are over virtual states 
only. Unlike fluorescence, the Raman effect is a light- scattering 
phenomenon, that is, the incident photon is not absorbed by the mole- 
cule. Instead, the molecule is perturbed so that a vibrational 
transition occurs. 5 

The physical meaning of these equations is that the intensities 
of Raman bands are governed by the relationship of the ground elec- 
tronic state to the excited states. If the ground state is vibroni- 
cally coupled to a single excited state, then only that state will 
influence the intensities of the Raman bands. Furthermore, only those 
vibrations mixing ground and excited states exhibit the resonance 
Raman effect. Likewise, if a pair of excited states are vibronically 
coupled, both states influence the Raman intensities. Once again, 



-27- 

only those vibrations mixing the two excited states exhibit the reso- 
nance Raman effect. 6 Experimentally, these can be distinguished by a 
study of relative intensities as a function of exciting frequency. 
Mathematically, the influence of a single excited state is related to 
diagonal vibronic energy terms whereas the influence of two excited 
states is related to off-diagonal vibronic energy terms. 7 Finally, 
only electronic transitions that are both spin-allowed and Laporte- 
allowed enjoy non-zero transition moments in the relevant equations, 
and so only these excited states can influence enhancement of 
intensities. 

Experimental 

The purposes of the resonance Raman technique are to study vi- 
brations which might otherwise have weak or badly- overlapped bands, 
to learn more about vibronic coupling of electronic states, and (in 
common with non- resonance Raman) to draw conclusions on the bonding 
within the molecule in question. Often reports have concentrated 
on this last aspect. For example, extensive studies have been con- 
ducted on tetrahedral, 8 octahedral, 9 and square planar 10 molecules 
incorporating halogens. From observed Intensities, it was determined 
that only single excited states governed the intensity enhancement in 
Group IV tetrahalides, but pairs of excited states influenced the in- 
tensity enhancement in square-planar and octahedral anions. Infor- 
mation on the bonding in these complexes was obtained through the 
calculation of mean bond polarizability derivatives, corrected for the 
resonance Raman effect. Results confirmed measurements by other tech- 
niques that the covalent character of the bond gradually increases as 
the halogen changes from fluorine to iodine. Also, covalent character 
was found to increase with increasing formal charge on the metal atom 
in anionic complexes. Mean bond polarizability derivatives could be 
decomposed into parallel and perpendicular bond polarizability deriva- 
tives. In all cases, the former proved to be more sensitive to the 
nature of the bond. Another study 11 on overtones of the permanganate 
and chromate ions allowed an estimate of bond dissociation energy to 
be made, indicated environmental effects on the widths of overtones, 
and suggested an equation predicting the relative intensity of the 
overtones. 

More complicated systems have generally been limited to biologi- 
cal molecules and their models, since resonance Raman provides a 
technique for monitoring vibrations of a chrornaphore independent of 
its matrix. 12 However, an investigation of 'cetracarbonyldi-|j-2, 2,5*5- 
tetramethylhex-3-yne-diiron has been reported. An analysis of relative 
intensities and depolarization ratios indicated that promotion of 
electrons in the iron-iron double bond is the transition which most 
strongly enhances intensity. A smaller contribution arises from an 
iron-acetylene charge transfer transition. And, in another investi- 
gation, 14 the resonance Raman spectrum of various ferricenium ions 
showed that only those vibrations involving the iron atom are enhanced 
enough to be observed. Two other bands are attributed to electronic 
transitions between low- lying spin-orbit states. 



-28- 

References 

1. M. C. Tobin, "Laser Raman Spectroscopy," Wiley-Interscience, 
New York, NY, 1971. 

2. J. Behringer, in "Raman Spectroscopy," H. A. Szymanski, Ed., 
Plenum Press, New York, NY, 1967, Chapter 6. 

3. J. Behringer, Z. Elektrochem. , 62, 906 (1958). 

4. A. L. Verma, R. Mendelsohn, and H. J. Bernstein, J. Chem. Phys., 
61, 383 (1974). 

5. L. A. Woodward, in "Raman Spectroscopy," H. A. Szymanski, Ed., 
Plenum Press, New York, NY, 1967, Chapter 1. 

6. A. C. Albrecht, J. Chem. Phys., 34, 1476 (1961). 

7. A. C. Albrecht and M. C. Hutley, J. Chem. Phys., 55, 4438 (1971). 

8. R. J. H. Clark and P. D. Mitchell, J. Mol. Spectrosc, 51, 458 
(1974). — 

9. Y. M. Bosworth and R. J. H. Clark, J. Chem. Soc, Dalton Trans. , 
1749 (1974). 

10. Y. M. Bosworth and R. J. H. Clark, Inorg. Chem., 14, 170 (1975). 

11. W. Kiefer and H. J. Bernstein, Mol. Phys., 23, 835 (1972). 

12. T. G. Spiro, Ace. Chem. Res., £, 339 (1974). 

13. G. J. Kubas and T. G. Spiro, Inorg. Chem., 12, 1797 (1973). 

14. B. P. Gachter, J. A. Koningstein, and V. T. Aleksanjan, 
J. Chem. Phys., to be published. 



-29- 

ELECTRON-NUCLEAR DOUBLE RESONANCE (ENDOR) OF 
TRANSITION METAL COMPLEXES 

Timothy R. Felthouse May 1, 1975 

Introduction 

The ENDOR method was proposed by Feher in lJ56 e ,as an extension 
of a dynamic nuclear polarization scheme. 1 Nuclear magnetic reso- 
nances at hyperfine frequencies were detected through their con- 
comitant effect on the electron paramagnetic resonance (EPR) signal. 
Applications of ENDOR to paramagnetic transition ions in diamagnetic 
hosts yielded nuclear g values and coupling constants for hyperfine 
and nuclear electric quadrupole interactions. The entire scope of 
materials subjected to the ENDOR technique has been diverse: para- 
magnetic defects in solids, organic radicals in solutions and liquid 
crystals, magnetically dilute metal complexes in single crystals and 
powders, and proteins. 2 " 6 

Theory and Experimental Approaches 

An ENDOR transition corresponds to a change in the nuclear mag- 
netic quantum number of one unit while the electronic quantum number 
remains fixed. Consequently, an ENDOR spectrum displays the EPR 
signal intensity as the nuclear radio frequency (in MHz) is swept. 
In order to account for the observability of an essentially nuclear 
transition, consideration must be made of the induced transition rates 
and spin populations between the electronic and nuclear levels, 
principal relaxation paths, and a radio frequency (rf) enhancement 
factor. 7 " 9 Three mechanisms can be distinguished to explain the full 
range of ENDOR effects: steady- state ENDOR, the "packet- shifting" 
model, and "distant ENDOR." The mechanisms are not mutually exclusive, 
although no comprehensive theory of the ENDOR mechanism in solids 
now exists. 

Four main directions have evolved by which the ENDOR experiment 
is performed: "low power ENDOR" in either the EPR dispersion of ab- 
sorption mode, transient ENDOR, and "high power ENDOR. 10 A com- 
mercially available spectrometer employs the "high power ENDOR" 
approach, using rf fields up to 12 gauss. 10 " 13 The intense nuclear 
fields are necessary for studies on solutions or solids with short 
relaxation times or transition metal complexes in which the delocal- 
ized spin density onto the ligands decreases the effect of the rf 
enhancement factor. 

Applications 

Rist and Hyde have investigated the ligand ENDOR of copper(Il) 
complexes in magnetically dilute organic single crystals and powders. 14 
In the single crystal work proton and nitrogen hyperfine couplings 
were determined as well as the nitrogen quadrupole coupling which is 
observed as a first-order splitting in the ENDOR spectrum. Proton 
resonances occur in pairs about the free proton frequency and are 
separated by their hyperfine interaction A, while a nitrogen ENDOR 
spectrum consists of four lines centered about A/2 with splitting 



-30- 



by the quadrupole and nuclear Zeeman interactions. A key feature 
which emerged out of the powder ENDOR work was that "single-crystal 
type" spectra could be obtained in some cases where there was a 
dominant magnetic interaction so that the magnetic field setting and 
modulation could select molecules in restricted orientations. How- 
ever, single-crystal ENDOR spectra are required for the assignment of 
resonances and hyperfine couplings to the ligand nuclei. 15 As an ex- 
tension of their work on copper(Il) complexes, Hyde and coworkers 
studied the ligand interactions in a frozen solution of the blue cop- 
per(Il) protein stellacyanin. l6 Although the ENDOR spectrum was not 
well resolved, they concluded that the copper ion was in a hydrophobic 
environment and at least one coordinating ligand was nitrogen. 

ENDOR has been quite successful in probing the various inter- 
actions in iron metalloproteins. Sands and coworkers determined the 
effective hyperfine tensors for the reduced two-iron ferredoxins. 17 ' 18 
For 57 Fe enriched samples the ENDOR line frequencies corresponded to 
a ground state of S=^> supporting the model of an antiferromagnetically 
coupled high-spin Fe(Il), Fe(lll) pair. Although the first investi- 
gation of single crystals of metmyoglobin was unable to resolve the 
ligand hyperfine couplings, 19 Feher and coworkers later found that 
frozen solutions enabled observation of proton, nitrogen, and 57 Fe 
hyperfine interactions. 20 Subsequently, mixed crystals of myoglobin 
containing about 90$ of the diamagnetic CO-liganded form have yielded 
a preliminary report of the diagonal components of the heme nitrogen 
hyperfine and quadrupole tensors and the 57 Fe anisotropic hyperfine 
interaction. 21 Further work on mixtures of hemoglobin and its met 
form hopes to monitor the cooperative oxygenation effect using ENDOR 
of various ligand nuclei. 



Hyperfine and quadrupole interactions that are normally too small 
for measurement from EPR spectra can be readily determined with high 
accuracy by the ENDOR technique. Furthermore, EPR spectra of trans- 
ition metal complexes often contain ambiguities in assignment of super- 
hyperfine splittings, whereas ENDOR spectra provide sets of resonances 
from magnetically equivalent nuclei. The ENDOR studies which have 
been performed on bioinorganic molecules should serve as a basis for 
applications to synthetic analogs and a variety of coordination 
compounds. 

References 

1. G. Feher, Phys. Rev., JL03, 834 (1956). 

2. A. L. Kwiram, Ann. Rev. Phys. Chem., 22, 133 (1971). 

3. K. Mobius and K. P. Dinse, Chimia, 2^7"46l (1972). 

4. L. R. Dalton and L. A. Dalton, Magn. Resonance Rev., 2, 36l (1973), 

5. N. M. Atherton, in "Electron Spin Resonance," Vol. 2, 

R. 0. C. Norman, Ed.., The Chemical Society, London, 197^* 
pp. 36-51. 



-31- 



6. J. S. Hyde, Ann. Rev. Phys. Chem. , 25, 407 (197*0. 

7. A. Abragam and B. Bleaney, "Electron Paramagnetic Resonance 
of Transition Ions," Clarendon Press, Oxford, 1970, Chapter 4. 

8. J. E. Wertz and J. R. Bolton, "Electron Spin Resonance: 
Elementary Theory and Practical Applications," McGraw-Hill 
Book Co., New York, N. Y. , 1972, Chapter 13. 

9. S. Geschwind, in "Hyperfine Interactions," A. J. Freeman and 
R. B. Frankel, Eds., Academic Press, New York, N. Y. , 1967, 
Chapter 6. 

10. Varian Instrument Division, "E-700 High Power ENDOR System," 
Varian Associates, Palo Alto, California, 1971. 

11. J. S. Hyde, J. Chem. Phys., 43, 1806 (1965). 

12. U. Ranon and J. S. Hyde, Phys. Rev., l4l, 259 (1966). 

13. J. S. Hyde, in "Magnetic Resonance in Biological Systems," 
A. Ehrenberg, B. G. Malmstrbm, and T. Vanngard, Eds., 
Pergamon Press, Oxford, 1967, pp. 63-84. 

14. G. H. Rist and J. S. Hyde, J. Chem. Phys., 49. 2449 (1968); 
ibid ., 50, 4532 (1969); ibid ., 52, 4633 (1970). 

15. A. SchweTger. G. Rist, and Hs. H. Gtinthard, Chem. Phys. Lett., 
31, 48 (1975). ' 

16. G. H. Rist, J. S. Hyde, and T. Vanngard, Froc. Nat. Acad. Sci. 
U. S., 67, 79 (1970). 

17. J. Fritz, R. Anderson, J. Fee, G. Palmer, R. H. Sands, 

J. C. M. Tsibris, I. C. Gunsalus, W. H. Orme- Johnson, and 
H. Beinert, Biochim. Biophys. Acta, 255 , 110 (1971). 

18. W. H. Orme- Johnson and R. H. Sands, in "Iron-Sulfur Proteins," 
Vol. II, Walter Lovenberg, Ed., Academic Press, New York, N. Y. , 
1973, Chapter 5. 

19. P. Eisenberger and P. S. Pershan, J. Chem. Phys., 47, 
3327 (1967). 

20. C. P. Scholes, R. A. Isaacson, and G. Feher, Biochim. Biophys. 
Acta, 263, 448 (1972); C. P. Scholes, R. A. Isaacson, 

T. Yonetani, and G. Feher, ibid ., 322, 457 (1973). 

21. G. Feher, R. A. Isaacson, C. P. Scholes, and R. Nagel, 
Ann. N. Y. Acad. Sci., 222, 86 (1973). 



-32- 

MECHANISMS OF SUBSTITITTION REACTIONS OF METAL CARBONYL HYDRIDE COMPLEXES 

Blaine H. Byers (Final Seminar) May 6, 1975 

Transition metal carbonyl hydrides of manganese and rhenium were 
first prepared in 1957 and 1958 respectively. Although numerous sub- 
stitution and exchange reactions have been examined, 1 * 2 the kinetics 
and mechanisms of these reactions have not been elucidated. 

Several different mechanisms for substitution reactions of vari- 
ous metal carbonyl complexes have previously been determined: CO 
dissociation for XM(COj 5 (X = halide); associative displacement for 
CoNO(CO) 3 ; and ligand migration for CH3Mn(C0) 5 . The emphasis of this 
study has been to determine which of these mechanisms obtains for the 
carbonyl hydrides. 

Substitution of PBu 3 for CO in HRe(CO) 5 has been investigated. 
The observed kinetics cannot be explained by any of the above mechan- 
isms. Rather, a unique radical chain process has been proposed. 
M(CO)s radicals (M = Re, Mn), which initiate the chain reaction, are 
generated by monochromatic irradiation of M 2 (CO)io at the wavelength 
corresponding to the metal-metal 0-0* transition. In the absence of 
M 2 (CO)io or other radical initiators (including trace impurities), 
substitution is extremely slow in hexane at ambient temperature. The 
following mechanism is consistent with these observations. 

hv 

Re 2 (CO) 10 5* 2Re(C0) 5 (l) 

Re(CO) 5 + L -> Re(CO) 4 L + CO (2) 

Re(CO) 4 L + L -» Re(CO) 3 L 2 + CO (3) 

Re(CO) 4 L + HRe(CO) 5 ■* HRe(CO) 4 L + Re(CO) 5 W 

Re(CO) 3 L 2 + HRe(CO) 5 "* HRe(CO) 3 L 2 + Re(CO) 5 (5) 

Re(CO) 5 _ x L x + Re(CO) 5 _ y * Re 2(CO) 10 _ x _ y L x+y (6) 

Reactions (2) through (5) are chain propagating steps, while step (6) 
terminates the process. Polysubstituted dirhenium species are ob- 
served as the reaction proceeds, further supporting the above mechan- 
ism. Extensions of this chain process to analogous halogen systems 
as well as polynuclear hydrides are likely. 

While attempting to clarify the above mechanism, the interaction 
of photochemically generated Re(CO)s radicals with molecular hydrogen 
was observed to produce several hydride species. At low photon flux, 
HRe 3 (CO)i 4 was produced, whereas higher flux irradiation yielded both 
H 2 Re 2 (CO) 8 and HRe(CO) 5 as the major species. Furthermore, kinetic 
examination of the substitution of PBu 3 for CO in Re(CO) 5 showed CO 
inhibition. Together, these observations support a mechanism involv- 
ing CO dissociation from Re(CO)s, oxidative addition of H 2 to the 15 
electron Re(CO) 4 moiety, abstraction of hydrogen by Re(CO) 5 > yielding 
HRe(CO) 4 . This species can then either dimerize to yield H 2 Re 2 (CO)e 
or interact with two molecules of Re(CO) 5 to give HRe 3 (CO)i 4 . Such 
activation of H 2 may have application to hydrogenation and hydro- 
formylation reactions. 



-33- 



Slmilar substitution reactions have also been studied for the 
more reactive HMn(CO)s. The experimental evidence suggests a combin- 
ation of mechanisms. At times, the reaction is extremely rapid, 
indicating a radical process. However, linear pseudo first order 
rate plots are observed when no radical process is involved, and a 
thermal reaction involving ligand dependent hydride migration appears 
to be operative: 

+L 

HMn(CO) 5 -* HCOMn(CO) 4 L ■> HMn(CO) 4 L + CO 

Although no spectroscopic evidence for the formyl species exists, 
13 C0 exchange studies,* currently being reexamined, do support such 
a pathway. 

This research suggests a correlation exists between M-H bond 
strength and the reaction pathway. The stronger Re-H bond shows no 
tendency to undergo hydride migration. Further studies in this area 
may allow one to predict the mechanism of substitution by knowing 
the approximate M-H bond strength. 

References 

1. H. D. Kaesz and R. B. Saillant, Chem. Rev., J2, 231 (1972). 

2. E. L. Muetterties, ed., Transition Metal Hydrides , 

Marcel Dekker, Inc., New York, 1971. 

3. F. Basolo and R. G. Pearson, Mechanisms of Inorganic Reactions , 

John Wiley and Sons, Inc., New York, second edition, 1967* 
Chapt. 7. 

4. A. Berry and T. L. Brown, J. Organomet. Chem., 33, C67 (1971). 



V/ TABI£ OF CONTENTS BBB'SmJS&Sn 



) 



INORGANIC SEMINAR ABSTRACTS 

1975.1976 am. jo Advaan she 

Summer Session : Page 

HIGH PRESSURE THERMODYNAMIC AND SPECTROSCOPIC STUDIES ON OXYGEN 1 

STABILIZATION AND COORDINATION - Tony J. Beugelsdijk 

SYNTHESIS, STRUCTURAL, MAGNETIC AND CHEMICAL PROPERTIES OF LOW 

VALENT ORGAN CMETALLIC TITANIUM COMPOUNDS - Dennis Sekutowski 3 

Fall Session : 

KINETICS AND MECHANISMS OF 13 CO EXCHANGE WITH Cr(CO) 4 (CHELATE) 
CCMPLEXES AND RELATED COMPOUNDS - Martin A. Cohen 1 

LAMELLAR COMPOUNDS OF GRAPHITE - Craig S. Chamberlain 2 

MULTIPLE METAL-METAL BONDING IN TRANSITION METAL DIMERS - 

Jack A. Kramer 4 

MAGNETIC CIRCULAR DICHROISM - Thomas C. Kuechler 7 

INTERNAL METALATION OF PHOSPHORUS LIGANDS - R. Joe Laws on 10 

METAL FLAVOENZYMES AND MODEL SYSTEMS - Karen Lynn Hassett 12 

I THE USE OF MATRIX ISOLATION SPECTROSCOPY IN THE CHARACTERIZATION 

OF DICOBALT OCTACARBONYL - Ray L. Sweany 14 

POSITRON ANNIHILATION SPECTROSCOPY - Bruce C. Bunker 16 

FREE RADICAL REACTIONS OF TRANSITION METAL ORGANOMETALLIC 

COMPOUNDS - Jerry Keister 19 

THE METAL BINDING SITES OF TRANSFERRIN - Steve Kessel 22 

REACTIONS OF COORDINATED NITROSYLS - Jan Peter Frick 26 

SOLID STATE PHOTOCHEMISTRY OF COORDINATION CCMPLEXES - John Breese 28 

TRANSITION METAL CATALYZED REARRANGEMENTS OF POLYCYCLIC 

HYDROCARBONS - Gordon F. Stuntz 30 

SCME ASPECTS OF THE OLEFIN METATHESIS REACTION - William Willis 33 

ONE DIMENSIONAL METALLIC CONDUCTORS DERIVED FROM THIRD ROW d 8 
TRANSITION METAL CCMPLEXES - Anton Elamma 36 



THERMODYNAMIC AND SPECTROSCOPIC STUDIES OF REACTIVITY AND 

BONDING IN TRANSITION METAL LEWIS ACID-BASE SYSTEMS - Marinda P. Li 38 



' 



Table of Contents (continued) Page 

ADDITION OF NUCLEOPHILES TO COORDINATED OLEFINS - Dennis Sepelak 40 

INORGANIC CHEMISTRY IN THE GAS PHASE AS REVEALED BY ION 

CYCLOTRON RESONANCE SPECTROSCOPY - J. Patrick Cannady 43 

SIGMA-PI REARRANGEMENTS IN ORGANOMETALLIC SYSTEMS - Keith Hodges 46 



-1- 



HIGH PRESSURE THERMODYNAMIC AND SPECTROSOPIC STUDIES 

ON OXYGEN STABILIZATION AND COORDINATION 

Tony J. Beugelsdijk 23 June 1975 

Introduction 

The function of molecular oxygen as a ligand in transition 
metal chemistry has evoked considerable interest within the 
past ten years. 1 The broad spectrum of included reactions 
is evidenced by the efforts of chemists from quite different 
interests. Biochemical interest focuses on the search for 
model compounds for oxygen transport, for oxidases, and for 
the catalytic insertion of oxygen atoms derived from molecular 
oxygen into biological substrates . 2 ' 3 ' 4 Industrial chemists 
search for homogeneous analogs to the current heterogeneous 
oxidation catalyst systems. 5 The scope of all such investigations, 
however, has reseluted in only a very incomplete definition 
of the nature of the chemical bond in terms of the thermodynamics 
of oxygen coordination, stabilization, and activation. 

Statement of the Problem 

A basic shortcoming in determining the thermodynamic 
parameters associated with oxygen uptake has been the choice 
of the experimental design. Spectroscopic titrations have 
been conducted over a limited fraction of the saturation range 
and experimental points fitted to a mathematical model which 
obscures non-ideal behavior and artificially smoothes observed 
uncertainties. 6 An alternate method, based on an esr integration 
technique suffers from errors inherent in such a manipulation 
and the delicate dependence of esr lineshapes on experimental 
conditions. 7 The ideal approach would see a maximum change 
in an observable per unit of perturbation on the system. In 
the absence of any new exotic methods, the remaining recourse 
is to force coverage of the saturation range under elevated 
pressures of oxygen. Accordingly, a high pressure optical 
absorption cell was constructed. 

Results and Discussion 

Toluene solutions of B-CoPIXDME show changes in the 
visible region accompanying increasing pressures of oxygen 
above solution characteristic of the formation of the dioxygen 
adduct (B=base, CoPIXDME= cobalt (II) protoporphyrin dimethyl 
ester) . The enthalpies of adduct formation show a remarkable 
constancy in light of the variation of base donor atoms studied. 

The failure to find a correlation between this enthaply 
and the base E and/or C property 8 may provide partial justification 
for the assumption of the presence of fTstabilization in the 
interaction. Attempts to conclusively defend such a mechanism 
failed due to the limited choice of bases with drastically 



-2- 



different E numhpr<5 v^a^ 

exert influent on ihereLUT^ S ?° Uld be e *P ecte ° to 
the base-metalloporphyrln adduct witf £ ° f the d or bltal s of 
degrees of overlap with thf orbTtal s ^ r $Ult of var y in 8 
changing overlap should M nffP«t 7* , ° f ox ygen. This 
the enthalpy of oxygen uptake lf to the "> a gmtude of 

species^v^wfwelYde^ne " "V* 6 "—ordinate 

and difffrence in the p^enUa^s 1 "?*^ T^ 8 ' ae Positions 

surprising constancy when ?he axiaf ^L* WO Waves show 

can be explained by an oxidati™ !J ^ Se was var ied. This 

lowed by a rapid intlrnal^^c^ron'translerf yrln nUCl6US fo1 ' 

" (B ' C ° (lI)(PO <lJf|l^ : MB : Co(III) ( Po r) 



(B-Co(II)(Por) 



-e: 



(B-Co(?)(p or )(0 2 ) 

sSlu%^r^LrL f o\Tda C ?L r n dinate Sp9CieS Was established in 

oxygen^^LL'tion^ul? a^i^r^ faCt ° rS underlying 
axial site. It is felt m! lar ?* r Perturbations at the 
exists between electronic and V sterio 8t * ^ ry delicate *«^nce 
perturbations may resu?rirLrever^ih^ U K r S mentS that such 
oxygen. & in lrre versible behavior toward 

References 

*' cite°d ffin! AdVan - Chem - Ser - 122, 111 (1971) and references 
" Rev"; 6|f t 269 r (l96i) M - Pal e e »°au ra , and S. E. Wiberly, Chem. 

I 0. Haya^f hi f C 5i^ yge ^ es ^^^ *> (1970). 

N. Y., (1962) uxygenases » Academic Press, New York, 

l' (Wi/^K 8nd °- J - °' C °" n °^ Coord. Chem. Rev., o, 145 

7 l : n^F-^o^^xSse^^fr 68 ' and b - *• james - 

7. F. A. Walker J. Ame?T Chem. hoi.] ,95 i 154 (1Q7 „ 
»• R. S. Drago, Struct. Bonding, 15, 73 (1973) . 3> ' 



-3- 

SYNTHESIS, STRUCTURAL, MAGNETIC AND CHEMICAL PROPERTIES OF LOW VALENT 

ORGAN (METALLIC TITANIUM COMPOUNDS 
Dennis Sekutowski FINAL SEMINAR June 30, 1975 

Low valent (less than iv) titanium organometallic compounds have 
been studied extensively by industry because of the economical impor- 
tance of Zlegler-Natta catalysts. Recently, other workers have become 
interested in biscyclopentadienyl titanium systems in relationship to 
nitrogen fixation and carbonyl insertion reactions. Despite all of 
this interest, there have been only a few low valent titanium systems 
that have been fully characterized. 

The reduction of Ti(iv) halides by various metals can lead to 
both Ti(lll) and Ti(ll) species. In the case of cyclopentadienyl de- 
rivatives, the Ti(lll) species are frequently complexed to the metal 
halide formed from the reducing metal, to give linear mixed metal com- 
plexes of the form (Cp 2 TiX) 2 MX 2 , where X = CI, Br and Mn = Zn, Be and 
Mn. 1 In addition to the oxidation potential of the central metal atom, 
the reduction reactions are very dependent upon solvent and the nature 
of substitution upon the cyclopentadienyl group. These complexes are 
extremely oxygen sensitive and the complex (Cp 2 TiCl) 2 ZnCl 2 can be used 
as an oxygen monitor for an inert atmosphere dry box. 2 

Only a few magnetic studies have been performed upon trimetallic 
systems and the magnetic susceptibility studies of the (Cp 2 TiX) 2 MX 2 
complexes revealed that these were the first examples of 1,3 magnetic 
exchange Tia a diamagnetic metal atom in a linear trimetallic com- 
plex. 3 * 4 The effects of variation in the cyclopentadienyl ring substi- 
tution, the bridging anion X, and the central metal M upon the 
cooperative interactions between the titanium atoms were investigated. 
The magnetic behavior observed for these species is anomalous when 
compared to that reported for the dinuclear (Cp 2 TiX) 2 complexes. 5 * 6 
A study of the magnetic and structural properties of (Cp^iClJg and 
(MeCp 2 TiCl) 2 were made in order to resolve the differences in the 
dinuclear and trinuclear systems. 7 

Another type of metal reduction product was observed in the 
reaction of (Cp 2 TiCl) 2 with sodium phenylethynyl. 8 Biscyclopenta- 
dienyltitanium phenylethynyl was previously assumed to be a dimer with 
the phenylethynyl groups bridging the metal atoms in either a o-tt 
fashion or with an electron deficient three center carbon bond. 9 The 
lack of a carbon-carbon triple bond stretch in the infrared spectrum 
and the diamagnetism of the compound do not agree with either of the 
above bonding descriptions. The structural work of [MeCp 2 TiC 2 C Q H5 ] 2 
indicates that the two phenylethynyl anions have oxidatively coupled 
and are complexed to two MeCp 2 Ti groups. The resulting complex is 
closely related to the uncoupled dimeric species and, on the basis of 
infrared studies, it is suggested that the latter is an intermediate 
in the coupling reaction. 



-4- 

References 

1. D. Sekutowski and G. D. Stucky, Inorg. Chem., in press. 

2. D. Sekutowski and G. D. Stucky, J. Chem. Ed., submitted for 

publication. 

3. R. Jungst, D. Sekutowski and G. D. Stucky, J. Amer. Chem. Soc., 

96, 8108 (197*0. 

4. D. Sekutowski, R. Jungst and G. D. Stucky, Ch. 11, 

Extended Interactions between Metal Ions in Transition Metal 
Complexes , L> V. Interrante, ACS Symposium Series, 5 (1974). 

5. R. S. P. Coutts, P. C. Wailes and R. I . Martin, J. Organometal. 

Chem., 47, 375 (1973). 

6. R. L. Martin and G. Winter, J. Chem. Soc, 4709 (1965). 

7. D. Sekutowslci, R. Jungst, J. Davis and G. D. Stucky, 

unpublished results. 

8. D. Sekutowski and G. D. Stucky, J. Amer. Chem. Soc, 

submitted for publication. 

9. J. H. Teuben and H. J. de LLefde, J. Organometal. Chem., 

17, 87 (I9 r ^>). 



-1- 

KINETICS AND MECHANISMS OF 13 CO EXCHANGE WITH Cr(CO) 4 (CHELATE) 

COMPLEXES AND RELATED COMPOUNDS 



Martin A. Cohen 



Final Seminar 



October 7, 1975 



Investigation of 13 C0 exchange with complexes of the type 
Cr(C0)4 (chelate) was undertaken to ascertain the rate ordering of 
ligands inducing cis labilization, and the stereochemistry of CO dis- 
sociation. Cis labilization is a phenomenon, previously noted by Brown 
and co-workers, 1 ' 2 in which a ligand coordinated to an octahedral metal 
center causes a rate enhancement of dissociation of the carbonyls cis 
to that ligand. On the basis of 13 C0 exchange studies and substitution 
studies, the proposed order of cis labilization is: 

Co, H" < P(0Ph) 3 < PPh 3 < I", py < Br" < CI" 

This research has involved study of the 13 C0 exchange of 
Cr(C0)4 (chelate) complexes, in which chelate = phen, bipy, diphos, dpp, 
ape, nbd, and cod. Three types of reactions were observed under the 
reaction conditions employed: a) decomposition; b) chelate loss with 
13 C0 incorporation into the metal fragment; and c) 13 C0 incorporation 
into the parent compound. It is proposed that, for the nbd and cod ? 
complexes, chelate loss occurs via the chelate ring-opening mechanism, 
i.e., a stepwise dissociation of the chelate ends. On the basis of the 
stereochemical distribution of 1 3 C0 in Cr(co) 4 ( 1 3 C0) 2 , it appears that 
the five-coordinate intermediate having the mono-ligated chelate is 
fluxional. Incorporation of 13 C0 into the parent complexes, containing 
phen, bipy, diphos, and dpp, occurs via CO dissociation cis to the che- 
late. The intermediate Cr(co) 3(L-L), generated by CO loss is partially 
fluxional, as evidenced by a comparison of the simulations of the pro- 
duct distribution for either a completely rigid or completely fluxional 
intermediate. 

An attempt to define further the labilizing effects of PPh 3 led to 
the exchange study of fac-Cr(co) 3 (phen) (PPh 3 ) and trans-Cr(co) 4 [PPh 3 ) 2 . 
In both cases, PPh 3 was dissociated. The products of the reaction [as 
determined by 1 3 C NMR) were fac-Cr(CO) si 1 3 C0) (phen) and a mixture of 
cis- and trans-Cr(co) 4 i 1 3 CO)PPh 3 , respectively. The explanation of the 
mixture of the cis and trans isomers is based upon the generation of a 
higher energy intermediate which is not readily accessible during the 
exchange study of Cr(co) 5 PPh 3 . Furthermore, the lifetime of the inter- 
mediate is of the same order as the rate of rearrangement to the more 
favored intermediate. 

References 

1. A. Berry and T. L. Brown, Inorg. Chem. , _11, II65 (1972). 

2. J. D. Atwood and T. L. Brown, J. Am. Chem. Soc, 97, 3380 (1975). 

3. G. R. Dobson, Inorg. Chem., 8, 90 (1969). 



LAMELLAR COMPOUNDS OF GRAPHITE 
Craig S. Chamberlain October 29, 1975 

Introduction 

Lamellar compounds of graphite (LCG) are composed of single 
layers of atoms or small molecules inserted between the parallel 
planar carbon layers of graphite. Inserted materials include 
alkali metals, transition metal halides and chalconides, non-metal 
halides and chalconides, halides, and acids. 1 Additionally, Klotz 
and Schneider 2 and Vol' pin and coworkers 3 have reported LCG with 
most of the first row transition metals in the zero-valent state. 

LCG have a variety of uses. They have been found to be syn- 
thetic metals, models for two dimensional magnetism, super- 
conductors, electrodes, redox agents, and catalysts. Included in 
the reactions that LCG catalyze are the Fischer-Tropsch synthesis 
of hydrocarbons and the synthesis of ammonia. 4 To better understand 
these compounds and how they function, it is important to know their 
structure and bonding. In this seminar, the structure and bonding 
in LCG of FeCl 3 , FeClo, Fe(o), and Mo(o) will be examined as repre- 
sentative cases, these compounds being the most well characterized in 
their respective classes. The primary physical methods used are 
X-ray diffraction, magnetism, Mossbauer spectroscopy, and electrical 
measurements. 

FeCl 3 and FeCl 2 LCG 

FeCls LCG are formed by heating graphite and allowing FeCl 3 
vapor to intercalate. 5 X-ray and electron diffraction studies indi- 
cate the FeCl 3 exists in the LCG as a distorted octahedron with some 
Cl-C interaction. 6 Mossbauer measurements tend to indicate that the 
FeCl 3 receives less than a complete electron from graphite upon LCG 
formation. 7 Thus, there is apparently no major perturbation of the 
FeCl 3 structure upon intercalation; the FeCl 3 is weakly held in place 
between carbon layers. 

FeCl 2 LCG are formed by reduction of the FeCi 3 LCG by heating in 
a stream of Np, Ar, or H P , or in a vacuum. Apparently, there is 
little or no bonding interaction between FeCl 2 and graphite as indi- 
cated by Mossbauer spectroscopy and the interlayer spacing determined 
from X-ray diffraction measurements. 

Fe(o) and Ho(o) LCG 

Fe(0) LCG may be formed by reduction of FeCl 3 LCG or FeCl 2 LCG 
with a variety of reducing agents such as Na-NH 3 , LiAlH 4 , or lithium 
biphenyl. 9 Mossbauer measurements indicate that the material formed 
is zero-valent iron which may be donating some electron density from 
^z orbitals to graphite. Magnetic data indicate that the iron may 
either be complexed or exist as a metallic monolayer. The spacing 
observed by X-ray diffraction Is in agreement with that expected for 



-3- 



van der Waals interaction between iron and carbon layers, indicating 
little bonding interaction. Thus, the question as to whether the 
iron is complexed or exists as a simple monolayer sandwiched between 
carbon layers is not entirely resolved. 3 

The bonding in the Mo(o) case is less ambiguous. This material 
is formed from M0CI5 LCG in the same manner as the Fe(o) LCG. 9 The 
compound is diama^netic, in agreement with the magnetic behavior ex- 
pected for a Mo(0; bis-arene complex, but not for a molybdenum mono- 
layer. The observed interlayer spacing is much smaller than expected 
for van der Waal separation of the layers or for olefinic or allylic 
complexes with graphite. The distance is in agreement with that ex- 
pected for a bis-arene complex of Mo(o). Thus, the Mo(o) LCG appears 
to be an example where graphite acts as a macroscopic bis-arenic or 
bis-hexahapto pi -donor ligand. 3 

References 

1. A. R. Ubbelohde and F. A. Lewis, "Graphite and Its Crystal 

Compounds," Clarendon Press, Oxford, England, i960. 

2. H. Klotz and A. Schneider, Naturwissenschafuen , 49 , 448 (1962). 

3. M. E. Vol' pin, et al. , J. Amer. Chem. Soc , 97, 3366 (1975). 

4. M. A. M. Boersma, Cat. Rev.-Sci. Eng . , 10, 243 (1974). 

5. H. Thiele, Z. anorg. allg. chem ., 207 , 3^0 (1932). 

6. J. M. Cowley and James A. Ibers, Acta Cryst . , 9, 421 (1956. 

7. A. G. Freeman, Chern. Comm ., 1968 , 193. 

8. J. G. Hooley, jet al., Carbon , 6, 68l (1968) and references 

therein. 

9. M. E. Vol'pin, V. I. Gol'danskii, et al., Bull. Acad. Sci . 

USSR Div. Chem. Sci ., 1970, 2452. 



-4- 

MULTIPLE METAL-METAL BONDING IN TRANSITION METAL DIMERS 

Jack A. Kramer October 30, 1975 

Introduction 

During the last decade, there has been a dramatic increase in 
the number of multiply-bonded transition metal compounds character- 
ized. Although the first reported synthesis 1 of a quadruply- bonded 
compound, Cr 2 (0 2 CCH 3 ) 4 (H 2 o) 2 , was in 1844, over 120 years passed 
before it was recognized 2 that such compounds are quadruply-bonded 
metal dimers. 

In 19^3, Russian workers reported 3 an unusual single-crystal 
X-ray structure for a compound with the empirical formula (pyH)HReCl 4 . 
The rhenium ions are separated by only 2.24 ft, and are positioned in 
the center of a nearly cubic array of chloride ions. This short Re-Re 
distance seemed quite unlikely in light of distances of 2.48 ft in 
Re 3 Cl 9 4 and 2.75 ft in rhenium metal. 

F. A. Cotton et al. 5 determined the crystal structure of 
"KReCl 4 •H 2 M , which confirmed the Russians' observations. He then 
proposed 2 a quadruple bond to account for the extremely short Re-Re 
distance and the eclipsed conformation of the chlorides. 

Occurrence 



In order to form metal-metal bonds, several conditions must be 
met; 6 the metal must a) be in a low oxidation state: high oxidation 
states lead to contracted d orbitals which don't allow the necessary 
overlap, and b) lie on the left of the periodic table. This prevents 
too large a contraction in the d orbitals and population of anti- 
bonding 'orbitals; 

Using these guidelines, a considerable number of compounds 7 
containing Re, Mo, Cr, Tc, Rh, Ru, Fe, and W have been found which 
have multiple metal-metal bonds. Mixed metal compounds containing 
Mo and W 8 and Cr and Mo 9 have been reported. 

Also, multiple metal-metal bonds are found in compounds with 
Br", CI", SCN , SeCN , -CH 3 , C 4 H 8 2 ",77'-allyl, 2 CR, P(C 2 H 5 ) 3 , 
PhC(NPh 2 ) 2 , and others as ligands. This shows the wide range of 
such compounds. 

Bonding 

It was proposed 2 that the quadruple bond has one c, two rr, and 
a 6 component. The a bond is formed by overlap of d 2 -p hybrids 
of each metal atom. The two equivalent v. bonds are formed from overlap 
of the degenerate d and d orbitals, and the 6 bond from overlap 
of the d orbitals. Only ¥he 6 bond is angle dependent, the others 
being cylindrically symmetrical about the metal-metal axis. The 
chlorides must be eclipsed for maximum overlap of the d orbitals. 
The s, p , p , and d 2 2 orbitals are used to bond to rne 
"equatorTal t,y ligands. " y 



The essentials of this bonding scheme have been verified by 
extended Huckel MO calculations 10 and SCF-XaSW calculations 11 on 
(Re 2 Cl 8 ) 2 ~, and SCF-XaSW calculations 12 on (Mo 2 Cl 8 ) 4 ~, although 
some details are still being debated. 

Many physical techniques have been used to probe the multiple 
metal-metal bonding, among them X-ray diffraction, 7 magnetic sus- 
ceptibility, 13 and photoelectron, 14 ~ 15 visible-UV, 16 epr, 16 nmr, 1T 
IR, ~ 21 and Raman 18 22 spectroscopies. 

Properties 

Compounds with multiple metal-metal bonds are generally air 
sensitive in solution, but relatively stable as solids. Solutions 
of Rh 2 4 " r (aq) are particularly stable, however, and may be handled 
in air. 23 

{Re 2 Cl 4 [P(C 2 H 5 ) 3 ] 4 } 2 may be reduced in solution to the +1 
and species, 24 but oxidation or reduction of most other compounds 
with multiple metal-metal bonds leads to unstable products. 

Protonation of Rh(ll), Ru(2.5), or Mo(ll) carboxylates and 
addition of triphenylphosphine forms complexes which catalyze hydro- 
formylation and carbonylation reactions, and the hydrogenation of 
olefins and acetylenes, either homogeneously or heterogeneously on 
cation-exchange materials. 23 

Summary 

Formation of multiple metal-metal bonds is an important aspect 
of the low- vale nt chemistry of many transition metals. They offer 
many opportunities to study metal-metal interactions as a function 
of the metals and ligands involved, as well as being useful catalysts 
for a number of reactions. There is much yet to be discovered about 
this unique class of compounds. 

References 

1. E. Peligot, Compt. rend., 19, 609 (l344). 

2. F. A. Cotton, Inorg. Chem., 4, 334 (1965). 

3. B. G. Kuznetzov and P. Koz'min, Zhur. Strukt. Khim., 4, 55 (1963). 

4. F. A. Cotton et al. , Inorg. Chem., 7 > 1563 (1968). 

5. F. A. Cotton and C. B. Harris, Inorg. Chem., 4. 330 (1965). 

6. F. A. Cotton, Acc'ts. Chem. Res., 2, 24 (1969). 

7. F. A. Cotton, Chem. Soc. Rev., 4, 27 (1975). 

8. R. E. McCarley et al., JACS, 97, 5300 (1975). 

9. C. D. Garner and R. G. Senior, J.C.S. Chem. Comm., 1974 , 580. 

10. F. A. Cotton and C. B. Harris, Inorg. Chem., 6, 924"Tl967) . 

11. H. B. Gray et al., Chem. Phys. Lett., 32, 283 (1975). 

12. J. Norman and H. Kolari, JACS, 97, 33"[l975). 

13. F. A. Cotton, Rev. Pure Appl. Chern., 17, 25 (1967). 

14. J. Green and A. Hayes, Chem. Phys. Lett., 31, 306 (1975). 

15. Walton, Inorg. Chem., 14, 2289 (1975). 






-o- 



16. Pettit, Quart. Rev., 2£, 1 (1971). 

17. J. San Filippo, Inorg. Chem., 11, 3l4o (1972). 

18. C. Oldham et al., Chem. Comm., 1971 , 572. 

19. C. Oldham et al., J.C.S. Dalton, 1975 , 23 04. 

20. F. A. Cotton et al., J. Coord. Chem., 1, 121 (1971). 

21. J. San Filippo et al., Inorg. Chem., 12, 2326 (1973) 

22. C. Oldham, J.C.S. Dalton, 1973 , 1067. 

23. G. Wilkinson, J. Chem. Soc. A, 1970 . 3322. 

24. F. A. Cotton et al., JACS, 97, 303 (1975). 



-7- 



MAGNETIC CIRCULAR DICHROISM 

Thomas C. Kuechler November 5* 1975 

Introduction 

Magnetic optical activity, discovered by Michael Faraday in 
1845, has proven to be a powerful technique in studying both the 
ground and excited electronic states of transition metal complexes. 
In this seminar, we are concerned with those aspects of magnetic opti- 
cal activity known as magnetic circular dichroism (MCD). Although the 
theoretical basis of MCD was available by 1965 1-3 and a few experi- 
ments had been done in the optical rotatory dispersion mode, major 
experimental advances did not come about until the introduction of com- 
mercial CD instrumentation and superconducting magnets in the mid- 
sixties. Thus, most experimental MCD is quite recent. A review of 
the theory and applications of MCD has recently been published. 4 



Theory 



In general, the form of an MCD line is given by: 

9,= - 21.346 (f°A + f'B + 4L£) 
m kT 



where 8 is the molar ellipticity, A, B and C are the Faraday terms 

and f°, m f are lineshape functions. For certain simplifying assumptions, 

the A,B, C terms are given by: 

A = -J- 2[<JlnJ J>-<alu la>][ |<almj J>| 2 - |<a|m_| j>| 2 ] 



2d 

a a, j 

B = ■■?- Z tZ v \ [<alm.| jXjIm |k>-<a|m | j><j|m Jk>] 
a a a,j k^a ^k ij a " " + 

+ Z E I [^Imj j><k|m | a>-<a|m |j><k|m + |j>] [ 
k^j k "j 

C=-«i- Z <aln _|a>[ |<a|m | |.i>| 2 - |<a|m|j>| 2 l 
a a, j 



Another term of frequent interest is the dipole strength: 



D = ^- ^ [|<a|m,),1>| 2 ■•- |<a|m_|j>| 2 ] 
a, j 



a 



An A term is possible, i.e., non-zero, whenever the ground or 
excited states of the transition are electronically degenerate. The 
magnetic field splits the degeneracy causing the right circularly 
polarized and left circularly polarized components of the transition 



-8- 

to occur at different energies. The C term arises when the ground 
state is degenerate. The magnetic field splits the degeneracy caus- 
ing a repopulation of levels according to Boltzmann statistics. The 
B term arises from field-induced mixing of the ground and excited 
states with other electronic states. This is the same effect that 
causes temperature independent paramagnetism. It is experimentally 
possible to separate an MCD spectrum into its component A,B,C terms 
if one measures the spectrum over a range of temperatures. 

Applications 

The simplest use of MCD lies in it being an alternate technique 
for detecting optical transitions. Relative intensities of bands in 
MCD are different from normal absorption spectroscopy making it pos- 
sible to detect some transitions better Ttfith MCD than with absorption. 
In particular, it is possible to detect spin-forbidden transitions 
with MCD which are often weak or even undetectable otherwise. 5 The 
resolution of bands in MCD is potentially better since both positive 
and negative bands occur. MCD has considerable value in distinguish- 
ing electronic from vibrational transitions in the IR since the latter 
show much weaker MCD. This is particularly applicable to transition 
metal complexes. 

MCD has found considerable use in assigning electronic transitions 
of transition metal complexes. The presence of an A or C term is im- 
mediate evidence that a degenerate state is involved. For example, 
Stillman and Thomson have shown that the presence of a negative A 
term in the MCD spectrum of Fe(ll)phthalocyanine in dichlorobenzene 
solution leads to the assignment of a 3 A 2f T ground state. 6 



o 



Most often the ground state is well-known, but the nature of the 
excited state is less certain. Here, comparison of experimental and 
theoretical values of A,B and/ or C allows the assignment of the cor- 
rect excited state to the transition. For example, Schatz and co- 
workers 7 have applied this technique to the charge-transfer bands of 
Fe(CN) 6 ~ 3 and Mn0 4 ~. From the sign of the observed C terms for 
Fe(CN) 6 ~ 3 > they were able to assign the symmetries of the excited 
states. For Mn0 4 , they were able to lend weight to one previous 
assignment and show that another assignment was incorrect. 

Once the assignment of a transition has been achieved, the use 
of experimental A values allows one to calculate excited state mag- 
netic moments. Such data is quite difficult to obtain by other means. 
Since the excited state magnetic moment is quite sensitive to Jahn- 
Teller effects, spin- orbit coupling and covalency effects, MCD is a 
sensitive probe of these effects. Stillman and Thomson have used MCD 
to measure the excited state magnetic moments of a series of metal 
phthalocyanines where the central metal was varied. 8 They ascribed 
the variations in the magnetic moment to various amounts of metal- 
ligand covalency. Similarly, McCaffery, Stephens and Schatz found 
small values of the excited state magnetic moment for several Co(lll) 
and Cr(lll) complexes and postulated a quenching of the orbital angular 
momentum in the excited states by Jahn-Teller interactions. 5 



-9- 

Additional applications include the elucidation of intensity 
mechanisms in the case of forbidden d-d transitions, 5 ' 9 and some work 
has been done in probing interactions between two metal centers. 10 * 12 - 
MCD has occasionally been used for identification of unknown species. 12 
As in absorption spectroscopy, the greatest amount of information is 
obtained from oriented samples at low temperatures where individual 
vibrational components are resolved. Such studies are becoming more 
prevalent. 13 

It is obvious that MCD is a powerful tool in studying both the 
ground and excited states of transition metal complexes of high sym- 
metry. MCD gives certain information more readily than other tech- 
niques and other data which are obtainable in no other way. MCD is a 
relatively new field experiencing rapid growth and promises to make 
significant contributions in the future. 

References 

1) P. N. Schatz and A. J. McCaffery, Quart . Rev ., 23, 552 (19&9). 

2) A. D. Buckingham and P. J. Stephens, Ann . Rev . Phys . Chem . , 17, 
399 (1966). 

3) P. J. Stephens, J. Chem . Phys . , 52, 3489 (1970). 

4) P. J. Stephens, Ann. Rev. Phys . Chem ., _25, 201 (1974). 

5) A. J. McCaffery, P. J. Stephens and P. N. Schatz, Inorg . Chem . , 
6, 16 14 (1967). 

M. J. Stillman and A. J. Thomson, J. Chem . Soc . Farad , II , 
1974, 790. 



6 

7 

8 

Q 

10 

11 

12 
13 



P. N. Schatz, A. J. McCaffery, W. Suetaka, G. N. Henning, 

A. B. Ritchie and P. J. Stephens, J. Chem . Phys . , 45, 722 (1966). 

M. J. Stillman and A. J. Thomson, J. Chem . Soc . Farad . II , 

1974 , 805. 

D. J. Robbins, M. J. Stillman and A. J. Thomson, 

J. Chem . Soc . Dalton , 1974 , 8l3. 

J. C. Cheng, A. Mann, G. A. Osborne, and P. J. Stephens, 

J. Chem . Phys . , 57, 4 051 (1972). 

A. J. McCaffery, J. A. Spencer and P. N. Schatz, Sym . Far . Soc . , 

3, 96 (1969). 

P. J. Stephens, Sym . Far . Soc , 3, 4o (19&9). 

R. W. Schwartz and P. N. Schatz, Phys . Rev . B. , 8, 3229 (1973). 






-10- 



INTERNAL METALATION OF PHOSPHORUS LIGANDS 

R. Joe Lav/son November 18, 1975 

Ligand and Metal Requirements 

Currently there is much interest in activating carbon hydrogen 
bonds. 1 Intramolecular activation of carbon hydrogen bonds occurs 
in transition metal complexes containing ligands coordinated through 
a Group V element. Only a few examples are known for arsines, 2 * 3 but 
much work has been done on systems containing amines, azobenzenes, 
phosphines and phosphites. 4-15 With aromatic amines and Schiff bases, 
the internal, or intramolecular, metalation is believed to take place 
by electrophilic substitution at the ortho ring position with the metal 
acting as the electrophile. In the case of phosphines and phosphites, 
however, many examples are thought to occur via oxidative addition in 
which the metal is nucleophilic. 4 

Internally metalated species have been isolated for all Group VIII 
metals except nickel, and species involving manganese have been re- 
ported. 4 " 8 The activated hydrogen is usually aromatic, but internal 
metalations of primary, secondary and tertiary carbons have been ob- 
served. 9 Metalation at an aromatic carbon is, however, much faster 
than at an aliphatic carbon. 1 

With phosphorus ligands, it is common to see either four or five 
membered rings formed, but in cases that both are possible, five 
membered rings are strongly favored. Thus, metalation involving com- 
plexes of P(CH 3 ) 2 (l-naphthyl) occurs exclusively at the 8 position of 
the naphthyl moiety to give five membered rings. 10 Bulky groups, such 
as tert-butyl, on phosphines have been shown to induce metalation even 
though the bulky group itself is not metalated. For example, trans- 
PtI 2 (PBu 2 Ph) 2 metalates at the ortho position of the phenyl ring, but 
cis-PtI 2 [p(CH 3 ) 2 Ph] 2 does not metalate. It has been proposed that this 
is due to an entropy effect. Since rotation is already hindered in the 
bulky phosphines, there is little entropy loss upon ring formation. 11 * 12 

Several dimetalated species have been isolated, usually with a 
third row transition metal. With iridium, two trimetalated species 
have been observed. 5 * 10 > 1 3 > 14 

Physical Studies 

Often the metalated and unmetalated species are in equilibrium 
under H 2 pressure. This can be easily observed by using D 2 and moni- 
toring the HD and H 2 formed or by checking the complex for deuterium 
incorporation. Deuterium incorporation has been used as evidence that 
metalation occurs, even if the metalated species cannot be isolated". 
Using RuHCl(PC 6 H 5 )3, one can catalytically ortho deuterate P(C 6 H 5 ) 3 . 
In coordinatively saturated complexes, which promote hydrogen deuterium 
exchange, addition of excess ligand slows or stops the exchange. 15 

Metalation at an aromatic carbon can be detected by ir; peaks 
corresponding to an ortho substituted benzene are observed. 13 The 
stereochemistry is usually determined by 31 P nmr spectra, although 



-11- 

in some hydrides only the 1 H nmr spectra with 31 P coupling have 
been used. 10 * 13 

Conclusions 

These studies have shown that metal complexes can aid in acti- 
vating carbon hydrogen bonds. It is hoped that continued investi- 
gations will lead to systems that catalyze substitution of aromatics, 
or even alkanes, by metalated intermediates. 

References ' 

1. G. W. Parshall, Accounts Chem. Res ., 8, 113 (1975). 

2. B. L. Shaw and R. E. Stainbank, jTc.S. Dalton , 2394 (1973). 

3. J. M. Duff, B. E. Mann, B. L. Shaw and B. Turtle, J.C.S. Dalton , 
139 (1974). 

4. G. ¥. Parshall, Accounts Chem. Res ., 3, 139 (1970). 

5. E. W. Ainscough, T. A. James, S. D. Robinson, and 
J. N. Wingfield, J.C.S. Dalton , 2384 (1974). 

6. E. K. Barefield and G. ¥. Parshall, Inorg. Chem ., 11, 964 (1972). 

7. L. K. Gosser, Inorg. Chem . , 14, 1453 (1975) . 

8. M. Y. Darensbourg, D. J. Darensbourg and D. Drew, 
J. Organometal. Chem ., 73, C25 (1974). 

9. D. F. Gill, B. E. Mann and B. L. Shaw, J.C.S. Dalton , 270 (1973). 

10. J. M. Duff and B. L. Shaw. J.C.S, Dalton , 2219 (1972). 

11. A. J. Cheney, B. E. Mann, B. L. Shaw and R. M. Slade, 
J. Chem. Soc. (a) , 3833 (l97l). 

12. A. A. Kiffen, C. Masters and L. Raynard, J.C.S. Dalton , 
853 (1975). 

13. E. W. Ainscough, S. D. Robinson and J. J. Levison, 
J. Chem. Soc. (a) , 3413 (l97l). 

14. A. J. Cheney and B. L. Shaw, J.C.S. Dalton , 754 (1972). 

15. G. W. Parshall, ¥. H. Knoth and R. A. Schunn, 
J. Amer. Chem. Soc, 91, 4990 (1969). 






-12- 



METAL PLAVOENZYMES AND MODEL SYSTEMS 



Karen Lynn Has sett 



November 25, 1975 



Molybdenum was the first metal discovered to be a component of 
a purified flavoenzyme, xanthine oxidase, in 1953. 1 In the inter- 
vening 22 years, there has been a considerable amount of interest in 
the exact role which metals play in enzymatic catalytic activity. 2 *" 



3.4 



Certain enzymes contain at least two types of prosthetic groups, 
or non-protein parts of the enzyme which are tightly bound to the pro- 
tein. These are the metalloflavoenzymes, which contain both a metal 
(Fe, Mo, and/or Zn) and a flavin. The two flavins commonly found are 
FAD (flavin adenine dinucleotide) and FMN (flavin mononucleotide). 
The structures of these two groups are as follows: 

R 


V FAD: R = CH 2 (CH0H) 3 CH2 0(P03H)2-adenosine 



H 




FMN: R = CH 2 (CH0H) 3 CH2 0P03H; 







Flavoenzymes have not proved easy to study. Their molecular 
weights are on the order of 300,000 g/mole, and they have, by defin- 
ition, at least two non-protein groups in addition to a long protein 
I chain. The most useful probe into the inner workings of these enzymes 
has thus far been electron paramagnetic resonance. Using this tech- 
nique, many different species containing unpaired electrons have been 
found. Molybdenum was one of the first to be identified definitely 5 
because of its characteristic six-line hyperfine splitting (25$ of 
natural abundance molybdenum has I = 5/2). As techniques of epr have 
improved, more has been discovered about the epr-active species in the 
enzymes. Not only is it known that Fe(lll) is involved in the cata- 
lytic activity, but there are two or four distinguishable iron sites 
in some enzymes. 6 ' 7 A note of caution is appropriate here, because 
although a good deal has been learned about the enzymes through the 
,use of epr spectroscopy, it is still true that no more than 20$ of the 
'total Mo, for example, has been accounted for in measurements on fully 
reduced enzymes. 8 If all of the molybdenum were participating simul- 
taneously in the oxidation of the substrate, 100!^ Mo would show epr 
signals (assuming that it is being reduced to the Mo(v) state, as is 
widely accepted for some Mo containing enzymes). 4 

Other detailed studies have seemingly elucidated the path that 
the electrons take during some oxidations; > 10 however, recent investi- 
gations have questioned the validity of some of these results. 11 

Since the detailed structure of these enzymes has thus far been 
too complicated for present techniques, many models which represent 
the prosthetic groups in the proteins have been made. The specific 
lodels under consideration here are the ones which mimic the possible 
lirect interaction between the flavin and metal. Two types of model 
systems have been looked at; those with a complete flavin structure, 



-13- 

such as riboflavin where the R in FAD or FMN is replaced by 
CH 2 (CHOH) 3 CH 2 OH, and also a series of 8 -hydroxy qui nones. The X-ray 
, structures, 1 2 * 13 electrical properties, 14 ' 15 and nuclear magnetic 
i resonance spectra 16 * 17 of these compounds have been studied 
extensively. 

REFERENCES 

1. E. De Renzo, E. Kaleita, P. Heytler, J. Oleson, B. Hutchings 
and J. Williams, J. Am. Chem. Soc , 75, 753 (1953). 

2. R. Bray and J. Swann, Struct . Bonding , 11 , 107 (1972). 

3. P. Hemmerich and J. Lauterwein" in "Inorganic Biochemistry," 
p. 1168, G. Eichhorn, Ed., Elsevier, New York, 1973. 

4. F. Bowden, in "Techniques and Topics in Bioinorganic Chemistry," 
p. 205, C McAuliffe, Ed., Wiley, New York, 1975. 

5. G. Palmer, R. Bray and H. Beinert, J. Biol. Chem., 239, 2657 
(1964). " 

6. D. Edmondson, D. Ballou, A. Van Heuvelen, G. Palmer and 
V. Massey, ibid., 248, 6135 (1973). 

7. T. Gray, P. Garland, D. Lowe and R. Bray, Biochem . J., 146 , 

239 (1975). 

8. H. Beinert and W. Orme- Johns on, in "Magnetic Resonance in 
Biological Systems," p. 221, A. Ehrenberg, B. Malmstrbm and 
T. V&nngfird, Eds., Pergamon Press, Oxford, 1967. 

9. K. Rajagopalan, P. Handler, G. Palmer and H. Beinert, 
J. Biol . Chem ., 243 , 3784 (1968). 

L0. K. Rajagopalan, P. Handler, G. Palmer and H. Beinert, ibid . , 
245 , 3797 (1968). 

111. R. Bray, D. Lowe and M. Barber, Biochem . J., l4l, 309 (1974). 
L2. W. Garland and C. Fritchie, J. Biol. Chem . , 2^9 7 2228 (1974), 

and ref. therein. 
L3. J. Gelder, J. Enemark, G. Wolterman, D. Boston and G. Haight, 

J. Amer. Chem . Soc . , 97, 1616 (1975), and ref. therein. 
L4. D. Sawyer, J. Gerber, L. Amos and L. De Hayes, J. Less - Comm . 

Metals , 36, 487 (1974). 
-5. A. Isbell and D. Sawyer, Inorg . Chem . , 10, 2449 (1971). 
.6. J. Lauterwein, P. Hemmerich and J. LHosTe. , ibid . , 14 , 

2152 (1975). 
.7. L. Amos and D. Sawyer, ibid., 13, 78 (1974). 



• 




-14- 

THE USE OF MATRIX ISOLATION SPECTROSCOPY IN THE 
CHARACTERIZATION OF DICOBALT OCTACARBONYL 

Ray L. Sweany December 11, 1975 

Currently two Isomers of dicobalt octacarbonyl are well 
enough characterized to have been accepted as real. Crystallographic 
studies have shown one of the isomers (I) to have two carbonyl 
ligands bridging the metals as shown below. A second isomer (II) 
is generally acknowledged to be of D symmetry with no bridging 
carbon monoxide. 2 Recently, it has been pointed out that the two 
isomers could not produce all the bands which have been observed in 
solution infrared spectra. A third isomer has been postulated. Raman 
evidence also supports the existence of a third isomer. 3 Observation of 
the 1 C nmr spectra have not been useful In studying the isomers of 
Co2(C0)g because of rapid interconversion between the isomers. 

I (Solid state 

solutions) II (Solutions) III (Proposed) 

The three isomers of Co2(CO)8 are easily differentiated using 
matrix isolation spectroscopy. Matrices of Co2(CO)8 are formed by 
co-depositing gaseous Co2*'CO)8 and a matrix gas on a sodium 
chloride support which is cooled by either liquid helium or nitrogen. 
The matrix gases used for this study were Ar, Ne, N2, 02, CO, CHi| , 
isobutane and hexane. The bands of the spectrum were assigned to 
their respective isomers by observing the intensity variation of 
the bands after the matrices were photolyzed by ultraviolet and 
visible radiation. There are three clearly differentiated sets of 
infrared bands, two of which correlate with the isomers discussed 
in the introductory paragraph. The third set of bands is assignable 
to a third isomer. We have tentatively assigned a structure of D2d 
symmetry to the third isomer (III, see above). 

The slow conversion of II to I has been observed in hexane 
matrices at 83°K. The rate constant for the transformation is 
2.4 + .2 x 10~^ s-1. The free energy of activation for the process 
is 6.3 + .4 kcal/mole. Isomer III is not observed in measurable 
concentrations in hexane even when the matrix is cooled with liquid 
helium. Its conversion to one of the other two isomers must be 
extremely rapid even at very low temperatures. An estimate of 
AG=: + 2.5 kcal/mole for such conversions seems consistent with the data. 

Photolysis of Co2(CO)8 yields new bands which are assigned 
to Co2(CO)7. The detailed structure of C02(CO)7 cannot be deduced 
from the spectrum. However, the absence of bands due to bridging 
carbonyls suggests the structure may be described by (CO) ijCo-Co(CO) 3 . 
Photolysis of Co2(CO)8 in CO matrices differs from the photolysis in 
other matrices. Co(C0)l| is observed when CO matrices are photolyzed 
as well as 002(00)7. 



-15- 

A mechanism which is consistent with the data appears below. 



Co 2 (CO)8 ^~±- Co 2 (CO) 7 + CO 



u./. 



2Co(C0) 



vi: 



TTV V 



t 



H "*T 



00(00)3 + Co(CO)i| + CO 



Pyrolysis of Co2(C0)q in the gas state yields a product tentatively 
identified as Co2(CO)6. The spectrum is consistent with a 
triple metal-metal bond. Interestingly, very little Co2(C0)y is 
observed under the same conditions. 

The results of the matrix experiments will be discussed in 
relation to the observed behavior of Co2(C0)g as described in 
several recent kinetic studies. 

References 

1. G. G. Sumner, H. P. Klug, L. E. Alexander, Acta Cyrstallog. , 
17, 732 (1964). 

2. G. Bor, K. Noack, J. Organometal . Chem. , 64, 367 (1974). 

3. S. Onaka, D. Shriver, in press. 



4. D. Kidd, D. Lichtenberger, P. Loeffler, T. L. Brown, unpublished 
data. 



-16- 

POSITRON ANNIHILATION SPECTROSCOPY 

Bruce C. Bunker January 20, 1976 

Implantation of positrons into solids, liquids, or gases results 
in the production of gamma radiation from the annihilation of the 
positrons with the electrons in the sample. The rate of annihilation 
and the energy and angular distribution of the gamma rays produced can 
provide a wealth of information concerning both the chemical and physi- 
cal environment of the electrons which are annihilated. To date, posi- 
tron annihilation spectroscopy has mainly been confined to solid state 
applications, but the chemistry which has been studied using this tech- 
nique indicates that it has great potential in chemical applications as 
well and could develop into as important a tool to the Inorganic chemist 
as Mbssbauer, ESCA, and other experimental techniques. 

Positrons are emitted from man-made nuclei deficient in neutrons 
such as uNa 22 with kinetic energies of around 1 MeV. They are rapidly 
thermalized in most materials and can then either annihilate with the 
electrons in the medium as free positrons or as positronium, which is 
an electron-positron bound pair. Ortho (triplet) positronium annihilates 
in around 10~ r sec. to produce three gamma rays. Para (singlet) posi- 
tronium and the majority of the free positrons annihilate via a two 
gamma process in around 10 10 sec. These annihilation times depend on 
the electron density of the medium in all cases. The lifetime of ortho 
positronium can also be shortened significantly by conversion to para 
positronium due to a collision with a paramagnetic center, or through a 
variety of chemical reactions such as oxidations, reductions, substi- 
tutions, and additions. Thus, much of the chemical information obtainable 
from a given sample is derived from a study of the annihilation of ortho 
positronium. 

Positron annihilation spectroscopy has to date involved four basic 
experimental techniques, all of which involve detection of the gamma 
radiation produced in the annihilation process. Rates of annihilation 
are measured using the lifetime experiment, which employes a positron 
source such as n Na 22 which emits a gamma ray at the same time it emits 
a positron. The time lag between the moment the "time marker" gamma ray 
is detected and the moment gamma rays produced by annihilation of the 
positron are detected gives the lifetime of the positron in the sample. 
Detection of changes in the relative amounts of ortho positronium formed 
is possible using a triple coincidence apparatus which detects all three 
gamma rays produced in the annihilation simultaneously. The momentum 
distribution of the electrons which are annihilated can be obtained 
either by doing an angular correlation experiment to measure the devi- 
ation from l8o in the angle between the gamma rays produced in the two 
gamma annihilation, or by analyzing the Doppler line shape of the 0.51 
MeV centered gamma annihilation band. Until recently, all of the above 
techniques were used independently of one another, but use of the tech- 
niques together should also prove to be of great value. 

Positron annihilation has a wide range of capabilities, most of which 
have not as yet been exploited to the extent that they should be. With 
this method, it is possible to detect the presence of free radicals and 
paramagnetic species with great sensitivity. Spin delocalization onto 



-17- 

the ligands of inorganic complexes can "be detected. Positronium re- 
actions can be used to determine the kinetic parameters for a variety 
of chemical reactions in solution. The method can be used to measure 
and identify defects and impurities in solids. Surface chemistry 
studies should also be possible. Phase transitions can be studied in 
detail by using positrons. Momentum distributions of the electrons 
in solids especially can be used in a variety of solid state and chemi- 
cal applications including determination of band structure and spin 
polarizations in metals and semiconducting materials, Compton scatter- 
ing applications, and others. The above list of applications is by no 
means complete and continues to grow as more investigations into the 
interactions between positrons and matter are carried out. Production 
of commercial instrumentation, improvements in instrument resolution, 
coupling of the various spectroscopic techniques, and development of new 
positron techniques employing such additions as magnetic fields are all 
needed if the method is to have the impact on the world of chemistry 
that it deserves. These coupled with a growing interest among chemists 
in the method should make positron annihilation spectroscopy one of 
the primary tools of the physical inorganic chemist within the next 
decade. 



-18- 

helY-rences 

1. J. Green ano J. Lee, Positroniuni Chemistry, (Acaaenic .fress, 
flew York, 1964). 

2. Werner Brandt, Appl. Phys., ^(l), 1 (1974). 

5. H. J. Ache, Angew. Chein. Int. hid., 11, 179 (±972) # 

4. V. I. Goldanskii, At. iinergj Rev., 6, 3 (1968). 

5. S. J. Tao and J. ri. Green, J. Ghem. Soc. A, 1968 , 408. 

6. W. J. Madia, A. L. Nichols, ana n, J. Ache, J. Am. Cnern. doc, 
2, 5041 (1975) . 

7. T. Chiba ana N. Tsuda, Appl. Phys., 5(1"), 37 (1974). 

8. H. M. tfingru ana P. K. Mijnarenas, Phys. Rev. B, ^(5) , 2372 (1974) . 

9. A. J. Nichols, it. E. Wild, L. J. Bartal, ana H. J. Acne, 
Appl. Phys., £(1), 37 (1974). 

10. V. I. Goldanszkij , B. i,evay, V. P. Santarovies, M. Ranogajec- 
iCo::ior ana Attila Vertes, i v ia^y. kern, Poly., bo, 490 (1974). 

11. o. is. Nicholas, R. ii. Wild, L. J. Bartal, ana H. J. Acne, 
J. Phys. Chem., 22, 178 (1973). 

12. R. jtfieminen, P. tfautojarvi, ana P. Ju&ho, Appl. Phys., £>(l), 41 (1974). 

13. H. J. Ache, L. J. Bartal, J. Inorg. Nucl. Chem., ^6, 267 (1974). 

14. V. I. Goldanskii, R. I. busman, Yu. W. violin, and V. P. tfhantarovich, 
i)okl. Akad. Kauk. JcjJR, English Trans., 188 , 660 (1969). 

15. A. Seeder, J. Phys. F, Metal Phys., Vol. 3, 248 (1973). 

16. Tumosa, tfang, anu Acne, J. Chein. Phys., £8(3)f 1833 (1^73)- 

17. Ashe, Angular Correlation of Annihilation Raaiation, Ph.D. Thesis, 
University of Texas, 1961 . 

18. Physica Status tfolidi B-, 6£(2), K105 (1975). 

19. T. L. Williams ana II. J. Ache, J. Chem. Phys., £0, 4493 (1969). 

20. L. J. isartax, J. B. itichoias,' ana xi. J. Ache, J. Phys. Chem., 
26, 1124 (1972) . 



-19- 

FREE RADICAL REACTIONS OF TRANSITION METAL ORGAN OMETALLIC COMPOUNDS 

Jerry Keister February 10, 1976 

Most mechanisms of transition metal organometallic reactions are 
based on simultaneous two- electron changes in oxidation state of dia- 
magnetic (usually 16- or l8-electron) systems. Exceptions have been 
redox reactions of organic radicals with metals which act as one-electron 
transfer agents (Pe(ll), Cr(ll), Cu(l), Co(ll)). The first compound for 
which a rich chemistry involving radicals was established was Co(CN) 5 ~ 3 , 
a 17-electron species which undergoes a variety of reactions, including 
oxidative addition of organic halides and hydrogenation of olefins, by 
radical processes. 1 Only recently have free radical intermediates been 
shown for other systems in reactions involving oxidative addition 2 ' 3 and 
reductive elimination, 15 hydrogenation, 14 insertion, 16 metal-metal bond 
formation and cleavage, 8 and substitution. 10 " 12 Studies of these radi- 
cal mechanisms have been greatly aided by knowledge about reactivity of 
organic radicals and by techniques developed in conjunction with the 
study of organic radicals. 

One of the most extensively studied reactions in organometallic 
chemistry is oxidative addition of alkyl and aryl halides to metal 
centers. This process has been shown to proceed by S N 2 displacement for 
many systems. 2 However, Osborn and coworkers 3 have shown that in certain 
cases this reaction can occur by a radical chain mechanism which is 
proposed to be: 

ML n + R- -> -MRL n 

•MRL n + RX -> XMRL n + R. 

This is indicated by the effects of free radical inhibitors and initi- 
ators on the rate of reaction. The reaction results in loss of stereo- 
chemistry at the carbon center, in contrast to an S N 2 type process, 
which gives inversion of configuration. In favorable cases, CIDNP 4 en- 
hanced absorption and emission, characteristic of free radical diffusive 
encounters, is noted in the nmr of the organic products. The presence 
of organic radical intermediates is also shown by their characteristic 
rearrangements and by hydrogen atom abstractions in good donor solvents. 
In other systems, free radical spin traps, such as t-butylnitroxide, 
have been used to trap the radical intermediates as relatively long- 
lived nitroxide radicals identifiable by esr spectroscopy. 5 However, 
the reaction mechanism is extremely dependent on the metal, its ligands, 
and the substrate. For example, Ir(C0)Cl(PMe 3 ) 2 reacts with BrCH 2 CHFPh 
by a free radical process but apparently by S N 2 displacement with CH 3 I. 

Another way in which metal-centered radicals can be generated is by 
homolytic cleavage of metal-metal bonds. In view of the radical chemis- 
try of Co(CN) 5 3 , it is perhaps not unexpected that its 17-electron metal 
carbonyl analogs (Co(C0) 4 , Mn(C0) 5 < Re(C0) 5 , Fe(C0) 3 (rT-C 3 H 5 ) , Mo(C0) 3 Cp) 
would behave similarly. The existence of such species is well estab- 
lished. 00(00)4 has been identified by its esr spectrum in a freshly 
sublimed sample of Co 2 (C0)e 6 and Re(C0) 3 (PMePh 2 ) 2 and other disubsti- 
tuted derivatives have been isolated and characterized. Wright on and 
others 8 have studied the photochemistry of metal dimers, including 



-20- 



M 2 (C0)io (M = Mn, Re) and [M(C0) 3 Cp] 2 (M = Mo, W) and substituted 
derivatives. Irradiation at the wavelength of the electronic transition 
assigned to the a -> a* transition of the metal-metal bond results in 
reversible homolytic cleavage. The radical nature of the products is 
revealed by their reactions with alkyl halides to produce the corre- 
sponding mononuclear metal halides and the organic radical. Lappert and 
coworkers 9 have succeeded in trapping and identifying the metal-centered 
radicals using large quantities of the spin trap nitrosodurene during 
photolysis of metal dimers. 

Recently, such 17-electron species have been proposed as inter- 
mediates in substitution reactions. Byers and Brown 10 studied the sub- 
stitution of HRe(C0) 5 with phosphines and proposed that Re(co) 5 is an 
intermediate in a radical chain substitution: 

•Re(C0) 5 + L -> -Re(C0) 4 L + CO 

♦Re(C0) 4 L + HRe(C0) 5 ~> HRe(C0) 4 L + -Re(C0) 5 

This reaction is inhibited by duroauinone and can be initiated by 
photolysis of small amounts of Re- 2 (co)in. In another study, these 
authors found that photolysis of Re 2 (COJi in the presence of PBu 3 forms 
Re 2 (C0) 3 L 2 and Re 2 (C0) 9 L, in addition to Re(C0) 3 L 2 , and, at constant 
light flux, the rate of substitution is independent of [PBu 3 ] and is 
inhibited by CO. 11 This was interpreted in terms of a dissociative 
substitution: 

L 
•Re(C0) 5 - -Re(C0) 4 + CO > Re(C0) 4 L 

In contrast, Poe, et al.. 12 have proposed that thermal substitution of 
Mn 2 (C0) 8 (PPn 3 ) 2 by P(0Ph) 3 to form Mn 2 (co) 8 (PPh 3 ) (P(OPh) 3 ) proceeds by 
an associative displacement: 

Mn 2 (C0) 8 (PPh 3 ) 2 * 2Mn(C0) 4 (PPh 3 ) 

Mn(C0) 4 (PPh 3 ) + P(0Ph) 3 -> Mn(co) 4 (P(OPh) 3 ) + PPh 3 

This was based on the increase in the rate of substitution with in- 
creasing [P(0Ph) 3 ] up to a limiting value that is identical to that for 
its reaction with CO. A 17-electron intermediate has also been shown 
for the reaction Co(CMe)| + TCNE ->_Co(CNMe) 4 (TCNE) + CNMe. 13 The 
intermediary of Co(CNMe) 5 2 and TCNE 7 has been demonstrated but the 
mechanism of substitution is not known. 

Metal-centered radicals have been implicated in the hydrogenation 
of polycyclic aromatic hydrocarbons by HCo(C0) 4 under "oxo" conditions. 14 
The mechanism proposed is: 



HCo(C0) 4 + C 14 Hio 
HCo(C0) 4 + -ChHh 



Co(C0) 4 + -C 14 Hn 
•Co(C0) 4 + C14H12 



The mechanism is similar to that proposed for the hydrogenation of 1,3- 
dienes by HCo(CN) 5 ~ 3 and accounts for the nearly equal amounts of cis 
and trans addition products observed and for the increase in rate with 
substitution at the hydrocarbon. 



-21- 

A variety of other reactions have been shown to involve free radi- 
cal intermediates but have not been investigated in depth. Cu(l) and 
Ag(l) alkyls in which there are no 3-hydrogens present decompose by 
homolytic cleavage of the metal-carbon bond. 15 Insertion of 2 into the 
cobalt-carbon bond of B 12 model compounds 16 and oxidation of zirconium 
alkyls to the alk oxides 1 7 appear to proceed via alkyl radical inter- 
mediates. Recently, the 17-electron Fe("nr-C 3 H 5 ) (C0)3 has been shown to 
be a catalyst for the isomerization of 1-hexene to trans -2-hexene with- 
out CO dissociation. 18 The mechanism is not known, but commonly proposed 
isomerization pathways are not completely satisfactory for this reaction. 
It is likely that further studies will show that radical processes for 
organometallic reactions are more common than has been believed. 

REFERENCES 

1. B. R. James, "Homogeneous Hydroge nation, " Wiley, New York, NY, 
1973, p. 106. 

2. J. Halpern, Accounts Chem. Res., 3, 386 (1970). 

3. (a) A. V. Kramer, J. A. Labinger, J. S. Bradley, J. A. Osborn, 

J. Am. Chem. Soc, 96, 7145 (197*0; (b) A. V. Kramer, J. A. Osborn, 

ibid., 9§, 7832 (197^); (c) J. S. Bradley, D. E. Connor, D. Dolphin, 

J. A. Labinger, J. A. Osborn, ibid., 94, 4o43 (1972); 

(d) J. A. Labinger, A. V. Kramer, J. A. Osborn, ibid., 95, 7908 

(1973). 

4. A. R. Lepley and G. L. Closs, "Chemically Induced Magnetic 
Polarization," Wiley, New York, NY, 1973. 

'a) M. F. Lappert, P. W. Lednor, J.C.S. Chem. Comm., 948 (1973); 

b) D. J. Cardin, M. F. Lappert, P. W. Lednor, ibid., 350 (1973); 

c) A. Johnson, R. J. Puddephatt , J.C.S. Dalton, 115 (1975). 

6. S. A. Fieldhouse. B. W. Fullam, G. W. Neilson, M. C. R. Symons, 
ibid., 567 (1974). 

7. J. T. Moelwyn- Hughes, A. W. B. Garner, N. Gordon, J. Organometal. 
Chem., 26, 373 (1971). 
(a) D. S. Ginley, M. S. Wrighton, J. Am. Chem. Soc, 97, 4908 

1975); (b) M. S. Wrighton, D. S. Ginley, ibid., 97, I F246 (1975); 

c) M. S. Wrighton, D. S. Ginley, ibid., 97, 2o65~jl975); 

d) J. L. Hughey, IV, C. R. Bock, T. J. Meyer, ibid., 97, 
Wo (1975). ~~ 

9. A. Hudson, M. F. Lappert, P. W. Lednor, B. K. Nicholson, 
J.C.S. Chem. Comm., 966 (1974). 

10. B. H. Byers, T. L. Brown, J. Am. Chem. Soc, 97, 949 (1975). 

11. B. H. Byers, T. L. Brown, ibid., 97, 3260 (1975). 

12. J. P. Fawcett, R. A. Jackson, A. Poe, J.C.S. Chem. Comm., 
733 (1975). 

13. R. C. Young, T. J. Meyer, D. G. Whitten, J. Am. Chem. Soc, 98, 
285 (1976). ~~ 

14. H. M. Feder, J. Halpern, ibid., 97, 7186 (1975). 

15. G. M. Whitesides, E. J. Panek, E. R. Stedronsky, ibid., 94, 
232 (1972). 

16. F. R. Jensen, R. C. Kiskis, ibid., 97, 5825 (1975). 

17. T. F. Blackburn, J. A. Labinger, J. Schwartz, Tet. Let., 3o4l (1975). 

18. E. L. Muetterties, B, A. Sosinsky, K. I. Zamaraev, J. Am. Chem. Soc, 
97, 5299 (1975). 



-22- 

THE METAL BINDING SITES OF TRANSFERRIN 

Steve Kessel February 24, 1976 

Introduction 

The existence of a unique iron bearing protein, human serum 
transferrin (Tf), was established in 19^7. 1} This iron binding pro- 
tein is but one of a family of such proteins, known as the trans- 
ferrins or siderophilins. As it does not contain a prosthetic group, 
such as heme, it must chelate the iron directly via the amino acid 
side chains. Serum transferrin serves several functions, the most im- 
portant of which is the transport of iron to the hemoglobin synthe- 
sizing reticulocytes. It also possesses effective iron buffering capa- 
bilities and bacteriostatic activity. 5 

The protein may be conveniently isolated by employing any two of 
the following techniques: alcohol fractionation, ammonium sulfate 
precipitation, ion exchange chromatography, electrophoresis, and 
gel filtration. 3 

General Physical Properties 

Early molecular weight determinations of transferrin based on 
ultracentrifugal measurements yielded values of 68,000 to 93,000 
daltons. More recent studies, though, have suggested a molecular 
weight of 76, 600. 4 Based on intrinsic viscosity measurements, the B- 
globulin appears to resemble a prolate ellipsoid. It is capable of 
specifically binding two atoms of iron per molecule. 3 > 3 > 6 

Chemical Composition and Structure 

The amino acid composition has been determined by several labora- 
tories. 3 > 7 Degradation of the protein into peptide fragments has been 
accomplished by enzymatic digestion and CNBr cleavage, but the complete 
sequence has not been determined. 8 > 9 The crystallographic unit cell 
parameters have been determined, but due to the difficulty in pro- 
curing satisfactory crystals, the detailed crystal structure has not 
been resolved. 10 

Transferrin appears to consist of a single polypeptide chain with 
no tendency to aggregate or dissociate in solution. 4 It should be 
pointed out that serum transferrin, as well as the other siderophilins, 
consists of approximately 5$ carbohydrate and hence is classified as a 
glycoprotein. 11 

Metal Binding Sites 

Serum transferrin reacts with iron in the presence of HC0 3 ~ to 
produce a red colored complex. The uptake of iron has been shown to* 
be stoichiometric, with two atoms of Fe specifically binding to each 
molecule of protein. 3 The overall reactions may be written in the 
following form : 5 > 6 



-23- 



HeTf 11 -!- Fe 5+ + HC0 3 " f=± [Fe-H 3 Tf -HCO3] n " 1 + 3K+ 
[Fe-HaTf-HCOa] 11 " 1 * Fe^\ HC0 3 " £=£ [(Fe ^-Tf-CHCOa);;,] 11 " 2 + 3H + 



K jFe) P -Tf-(HC0 n ) P ] Q - 2 [H^ 
* 2 [Fe-HaTf-HCOs] 11 -- 1 - (HC0 3 "] [Fe^ 



^F 



The prevalent view for some time was that Fe 3 ions went on and off 
the apoprotein two at a time. More recent experiments have indicated 
otherwise, that apotransferrin, monoferric transferrin, and diferric 
transferrin must all simultaneously exist. 15 > 16 The formation constants 
are quite high (^10 30 ) and appear to be almost equivalent. 17 

A number of multivalent metallic ions will complex with Tf . 
Fe(lll), Zn(ll), Cr(lll), Co(ll), Cu(ll), Mn(ll), and many others, in- 
cluding certain rare earth metals, are tightly and specifically bound 
by Tf. 3 Some of the metals have been ordered in terms of stability 
of metal-protein complex by displacement studies. 6 

Fe(Hl)>Cr(lIl), Cu(ll)>Mn(ll), Co(ll), Cd(ll)>Zn(ll)>Ni(ll) 

The nature of the reacting groups involved in the binding of the 
metals has been studied by a variety of means, including chemical modi- 
fication of amino acid residues, spectrophotometric titration, visible, 
ultraviolet, infrared, fluorescence, ORD, CD, NMR, and ESR spectroscopy. 
Overall, the evidence indicates that three tyrosyl residues, two to 
four nitrogen ligands (presumably histidine), and one bicarbonate ion 
are involved in the binding of a metal ion at each site. 

Tf shows two intense absorption bands in the ultraviolet and 
visible region. The^red ferric transferrin complex exhibits absorption 
maxima at 280 nm (E^L = 14.3) and 468 nm (E^' = 0.57). 3 The origin 
of these bands is stITI not clearly understood. The visible band may 
be tentatively assigned to a phenolate >Fe(lIl) , pn-dn* charge transfer 
transition. 12 As illustrated by optical rotary dispersion and circular 
dichroism studies, both absorptions are optically active. 13 * 14 

Magnetic susceptibility data has shown that the bound iron is in 
the high spin ferric state. 18 * 19 > 14 Little, if any, interaction tran- 
spires between the bound metal ions, as depicted by the static magnetic 
susceptibility. The ESR spectrum of the iron-bicarbonate-transferrin 
complex is characteristic of the class of Fe(lll) compounds in which 
iron is in a site of rhombic symmetry. It consists of a three component 
almost isotropic line-shape centered around g' =4.3 and two less in- 
tense anisotropic transitions around g' = 9 and g' = 1,5,20,21,15 
Attempts have been made to evaluate the spectrum by theoretical calcu- 
lations, using the spin-Hamiltonian 



V 



/< 



gBH-S + D[S 2 -l/3S(S-hl)] + E(S 2 -S 2 ). 22 ' 23 
^ x y 



-24- 

The primary feature of the spectrum, which occurs in the region around 
g f =4.3, arises from transitions within the middle Kramer's doublet. 
A zero field splitting value of 1.6 + 0.01 °K, with D = 0. 32 cm" 1 and 
X = 0.315 best explains the experimental data. 20 

While it was believed for some time that Tf would bind iron even 
in the absence of bicarbonate, 3 > 24 it is now reasonably certain that 
HC03~ or some other suitable anion is required for complete complex- 
ation.^ ESR spectroscopy has revealed that while HC03~ is the preferred 
anion. 25 EDTA, nitrilotriacetate, oxalate, 24 malonate, and thioglycol- 
late 25 are all capable of inducing chelation of iron by Tf . The question 
as to the equivalency or nonequivalency of the two binding sites has 
also been cleared up by use of ESR. Replacement studies of Cr(lll) bound 
to Tf clearly indicate a difference in the ligand field imposed by the 
two sites. 14 Evidence that Fe(lll) is also bound by nonidentical sites 
has been provided by ESR saturation techniques. Yet further evidence 
corroborating this has been presented by studying the ESR spectrum of 
iron-bicarbonate-transferrin as a function of perchlorate concentration. 26 

The ESR spectrum of Cu(ll) bound to Tf is indicative of a spin= 1/2 
system in an axial environment. 15 The substantial decrease in zero field 
splitting of Cu(ll) compared to Fe(lll) is manifested in the appearance 
of superhyperfine splitting in the copper spectrum, implying the presence 
of nitrogen ligands. 21 

REFERENCES 

1. C. G. Holmberg and C. B. Laurell, Acta Chem. Scand., 1, 944 (1947). 

2. A. L. Schade and L. Caroline, Science, 104, 3420 (19467. 

3. P. Aisen, "The Transferrins (Siderophilins), " Inorganic Biochemistry , 
ed., Gunther L. Eichhorn, Elsevier, New York (1973). 

4. K. G. Mann. W. W. Fish, A. C. Cox and C. Tanford, Biochemistry, 9, 
1348 (1970). 

5. E. R. Giblett, "F. Transferrin," Physiological Pharmacology , 
eda, Walter S. Root and Nathaniel I. Berlin, Academic Press, 
New York (1974). 

6. E. H. Morgan, "Transferrin and Transferrin Iron," Iron in 
Biochemistry and Medicine , eds., A. Jacobs and M. Worwood, 
Academic Press, New York (1974). 

7. A. Bezkorovainy and R. H. Zschocke, Arzneim. -Forsch. , 24 , 
476 (1974). 

8. P. Charet, C. R. Acad. Sci., Ser. D, 28o, 2049 (1975). 

9. M. R. Sutton. R. T. MacGilliviary, and K. Brew, Eur. J. Biochem., 
51, 43 (1975). 

10. B. Magdoff-Fairchild and B. W. Low, Arch. Biochem. Biophys., 158 , 
703 (1970). 

11. H. E. Sutton and G. A. Jamieson, "Transferrins, Haptoglobin and 
Ceruloplasmin, " Glycoproteins , ed. Alfred Gottschalk, Elsevier, 
New York (1972). 

12. B. P. Gaber, V. Miskowski, and T. G. Spiro, J. Amer. Chem. Soc, 
96, 6868 (1974). 

13. R. E. Feeney and St. K. Komatsu, Structure and Bonding, 1, 149 
(1966). 



-25- 



P. Aisen, R. Aasa, and A. G. Redfield, J. Biol. Chem., 244, 

4628 (1969). 

R. Aasa, B. G. Malmstrom, P. Saltman, and T. Vanngard, 

Biochim. Biophys. Acta , 75, 203 (1963). 

J. V. Princiotto and E. J. Zapolski, Nature, _255, 87 (1975). 

P. Aisen and A. Leibman, Biochim. Biophys. Res. Commun. , 52 , 

220 (1968). 

A. Blaise and J. L. Giradet, C. R. Acad. Sci., Ser. D, 269 , 

966 (1969). 

A. Ehrenberg and C. B. Laurell, Acta Chem. Scand., 9, 68 (1955) 

P. Aisen and R. A. Pinkowitz, J. Biol. Chem., 247 , 7^50 (1972). 

R. Aasa and P. Aisen, J. Biol. Chem., 245 , 259"9~Tl968) . 

R. Aasa, J. Chem. Phys., 52, 5919 (197017 

R. D. Dowsing and J. P. Gibson, J. Chem. Phys., _50, 294 (1969). 

P. Aisen, R. Aasa, B. G. Malmstrom, and T. Vanngard, 

J. Biol. Chem., 242, 2484 (1967). 

P. Aisen, A. Leibman, R. A. Pinkowitz, and S. Pollack, 

Biochemistry, 12, 5679 (1975). 

J. F. Gibson and E. M. Price, J. Biol. Chem., 247, 8051 (1972). 



-26- 

REACTIONS OF COORDINATED NITROSYLS 

Jan Peter Frick March 11, 1976 

Interest in the reactions of nitric oxide coordinated to trans- 
ition metals has increased in the last few years. 1 This increase is 
partly due to the discovery by D. J. Hodgson and J. A. Ibers that 
IrCl(C0)(N0)(p(C 6 H 5 )3)2BF 4 has an Ir-N-0 bond angle of 124°. This dis- 
covery led to a number of crystal structures that showed that the 
M-N-0 angles can vary from 120° to l8o°. The different types of nitro- 
syl bonding are shown below: 





III 


:0: 
II 


'6 


M M 


M M 


• • 

3 

111 


N 
1 


N 
1 


■N 

1 


N 
II 


N 
II 


^ N M-N-O-M' 4 > 5 
M— I — M 


M: 


M 


M 


•0» 

• • 


:0: 


^M^ 



The bond angle cannot be determined by N-0 infrared stretching 
frequency. The bent nitrosyls, with angles between 120° and 130°, have 
stretching frequencies between 1500 cm" 1 and 1720 cm" 1 . For the linear 
nitrosyls {170° to l8o°), the stretching frequencies vary from 1600 cm" 1 
to 1950 cm 1 . The lack of correlation between N-0 stretching frequency 
and bond angle, means that crystal structure is the only way to deter- 
mine the M-N-0 bond angle. Several attempts have been made to predict 
the bonding angle. They are based on the reducing power of the metal, 
charge of the complex, coordination number, other ligands present, 
molecular orbital calculations and symmetry. e > 7 > B ) 9 

The interest in the angle of the M-N-0 bond arises because the 
two types react differently. The linear nitrosyl reacts like NO and 
undergoes nucleophilic attack. An example of this is attack by hy- 
droxide. The overall reaction uses two hydroxides and forms a nitrite. 
The proposed mechanism is: 

Fe(CN) 5 N0" 2 + OH" -> Fe(CN) 5 N" ° H - 3 

Fe(CN) 5 N^Q H -3 + OH" ~> Fe(CN) 5 N0 2 ~ 4 + H 2 x 

Other nucleophilic attacks occur with N 3 ", 10 ^ X1 NH 3 , 12 N 2 H4, 12 * 13 
Cr +2 , 14 Zn 15 and H 2 NAr. 16 > 17 

Bent nitrosyls react like N0~. An example of an electrophilic 
attack is dioxygen on Co(salen)N0. l8 The cobalt complex needs a base 
such as pyridine or phosphine in the sixth coordination site for di- 
oxygen to react. The proposed mechanism is: 

Co(salen)N0 + B ^ BCo(salen)N0 

BCo(salen)N0 + 2 -» BCo(salen)N^°_ 

BCo(salen)N^Q_ + BCo(salen)N0 -> BCo(salen)N^Q_Q>CoB(salen) 

BCo(salen)N^Q_Q;NCoB(salen) -> 2 BCo(salen)N0 2 



-27- 

Other reactions of the bent nitrosyls are with NO, 19 and H + . 20 > 21 
An interesting reaction was first observed as: 

Ir(N0) 2 (P(C 6 H 5 ) 3 )2 + CO -> Ir(C0) 3 (P(C 6 H 5 )3)2 

This apparent substitution was unexpected since nitrosyls are not re- 
placed by carbon monoxide. This is due to the strong M-N bond found 
in transition metal nitrosyls. As the reaction was studied, it was 
found that the other products were nitrous oxide and carbon dioxide. 
This meant the overall reaction was: 

Ir(NO) 2 (P(C 6 H 5 ) 3 ) 2 + ^ CO ■* Ir(CO) 3 (P(C 6 H 5 ) 3 ) 2 + N 2 + C0 2 22 

A recent study on a Rh analog found that water and acid increased the 
rate of reaction. 23 Due to environmental concerns over the production 
of nitric oxide, new nitrosyl complexes will be studied with the hope 
of finding an efficient catalyst for the decomposition of nitric oxide. 

References 

1. N. G. Connelly, Inorg. Chim. Acta Rev., 6, 47 (1972). 

2. D. J. Hodgson and J. A. Ibers, Inorg. Chem. , J_, 2345 (1968). 

3. R. C. Elder, Inorg. Chem., 13, 1037 (1974). 

4. D. B. Brown, Inorg. Chem., ITjT, 2582 (1975). 

5. A. E. Crease and P. Legzdins, J. Chem. Soc. Dalton Trans., 1501 
(1974). 

6. B. L. Haymore and J. A. Ibers, Inorg. Chem., 14, 2610 (1975). 

7. B. L. Haymore and J. A. Ibers, Inorg. Chem., 17[, 3o6o (1975). 

8. R. Hoffmann, M. M. L. Chen, M. Elian, A. R. Rossi and 
D. M. P. Mingos, Inorg. Chem., 13, 2666 (1974). 

9. J. H. Enemark and R. D. Feltham, Coord. Chem. Rev., 13* 339 (1974). 

10. S. A. Adeyemi, F. J. Miller and T. J. Meyer, Inorg. Chem., 11, 
994 (1972). 

11. P. G. Douglas and R. D. Feltham, JACS, 94, 5254 (1972). 

12. F. B. Bottomley and J. R. Crawford, Chem. Comm. , 200 (1971). 

13. P. G. Douglas, R. D. Feltham and H. G. Metzger, JACS, 93, 84 (l97l) 

14. J. N. Armor, M. Buchbinder and R. Cheney, Inorg. Chem., 15 , 
2990 (1974). 

15. F. Bottomley and S. Tong, Inorg. Chem., 13, 243 (1973). 

16. W. L. Bowden, F. L. Little and T. J. Meyer, JACS, 95, 5084 (1973). 

17. W. L. Bowden, F. L. Little and T. J. Meyer, JACS, 9H, 444 (1976). 

18. S. C. Clarkson and F. Basolo, Inorg. Chem., \2, 1528 (1973). 

19. W. B. Hughes, Chem. Comm., 200 (1969). 

20. G. Dolcetti, N. ¥. Hoffman and J. P. Collman, Inorg. Chim. Acta, 
6, 531 (1972). 

21. K. R. Grundy, C. A. Reed and W. R. Roper, Chem. Comm., 1501 (1970). 

22. B. F. G. Johnson and S. Bhaduri, Chem. Comm., 650 (1973). 

23. R. Eisenberg and C. D. Meyer, Acct. Chem. Res., 8, 26 (1975). 



SOLID STATE PHOTOCHEMISTRY OF COORDINATION COMPLEXES 

John Breese | March 23, 1976 

INTRODUCTION 

During the past 20 years an increasingly large number of photo- 
chemical investigations of coordination complexes in solution have 
been reported. 1 However, in spite of this proliferation of photo- 
chemical studies of these species in solution, the photochemistry of 
solids has developed at a very slow rate. Although many reasons can 
be advanced for this slow development, including, until recently, a 
general lack of interest among inorganic photochemists, the major 
problem has been experimental. Thus, "although the inter- and intra- 
molecular forces of an ordered periodic crystal lattice can be treated 
theoretically with considerably more certainty than the constantly 
changing conditions of fluid solutions, the analytical tasks, so neces- 
sary for solution of even simple reaction schemes, are unreasonably 
complicated because the rigid structures prevent facile separation of 
reactants and products." 2 However, because there are important funda- 
mental concepts with roots in solid systems, the photochemistry of 
inorganic systems in the solid state is an area worthy of closer 
attention by Inorganic Chemists. 

This discussion will be limited to the so-called "robust" coordin- 
ation complexes, that is, complexes that maintain their coordination 
sphere when dissolved in fluid solution. The attempt will be to illus- 
trate, from the small amount of data available, situations in the 
photochemistry of solids that are analogous to those of solutions and 
then to point out the properties of solids that can modify their 
behavior. 

EXPERIMENTAL TECHNIQUES 

When ligand field bands of a solid are photochemically excited, 
99$ of the light incident on the solid surface (assuming no reflection 
loss) is absorbed in a region 5-10 micrometers thick. If the excite- 
ment occurs in a charge* transfer region, the absorption layer is even 
thinner. It is the miniscule nxture of this reaction zone that makes 
analysis of photoproducts very difficult and detection of reaction 
intermediates virtually impossible. 

Two methods which are most popular for identifying and analyzing 
the photoproducts of coordination complexes have been the measurement 
of diffuse reflectance spectra 3 and the infrared spectrum of materials 
dispersed in alkali halide pellets. 4 Both techniques are well suited 
to the qualitative observation of the photochemistry of solids because 
they are sensitive to the changes on surfaces. Unfortunately, it is 
extremely tedious to calibrate these methods for the determination of 
quantum yields and therefore few such results appear in the literature. 

The most common method used for the determination of quantum 
efficiencies for solid state reactions involves the use of reaction- 
rate plots from optical transmittance data for smooth layers or plates 



-29- 

of material (such as a pressed alkali halide pellet). 5 The method 
combines the Beer-Lambert equation for light absorption with the 
appropriate reaction- rate equation. This gives an exact solution 
to the problem for the case in which the photoproducts are trans- 
parent. Unfortunately, for reactions of coordination complexes, the 
products often absorb as efficiently as the reactants. In this case, 
a similar treatment yields an approximate equation which is valid 
for short illumination time. 6 

PHOTOREACTIONS 

The absorption spectra of solid coordination complexes are funda- 
mentally the same as for the complexes in solution. 7 There are, 
however, some differences in the subsequent thermal (or "dark") re- 
actions as the emission spectra of some metal complexes (e.g., with 
Cr 3 ) are different in the solid state than in solution. Still, the 
photochemical reactions in solids are to a large extent similar to 
those that are observed in solution. 8 ' 9 That is, phot ©substitution, 10 
charge-transfer photoredox, 11 and linkage isomerization 12 are observed 
in both solution and the solid state (photoracemization which is known 
to occur in solution has not yet been observed in the solid state). 
However, because of the rigidity of the solids and their periodicity, 
unusual adaptations of these reactions can be observed. 13 

REFERENCES 

1. V. Balzani and V. Carassiti, Photochemistry of Coordination 
Compounds , Academic Press, New York, 1970. 

2. Paul D. Fleischauer, in Concepts of Inorganic Photochemistry 
(Arthur W. Adams on and Paul D. Fleischauer, Ed.), Chapt. 9> 
Wiley Interscience, New York, 1975. 

3. W. W. Wendlandt, in Analytical Photochemistry and Photochemical 
Analysis (J. M. Fitzgerald, Ed.), Chapt. b, Marcel Dekker, 

New York, 1971. 

4. G. Lohmiller and ¥. W. Wendlandt, Anal. Chim. Acta., 51 , 
117 (1970). 

5. E. L. Simmons and W. W. Wendlandt, Coord. Chem. Rev., 7, 11 (1971). 

6. E. L. Simmons, J. Phys. Chem., 75, 588 (1971); E. L. Simmons and 
W. W. Wendlandt, Anal. Chim. Acta., 53, 8l (1971). 

7. L. S. Forster, in Transition Metal Chemistry (R. L. Carlin, Ed.), 
Vol. V, pg. 1, Marcel Dekker, New York, 1969. 

8. See Chapt. 18 in reference 1. 

9. P. D. Fleischauer and P. Fleischauer, Chem. Rev., 70, 199 (1970). 

10. M. R. Snow and R. F. Boomsma, Acta Crystallogr. B. , _28, 1908 
(1972); C. H. Stembridge and W. W. Wendlandt, J. Inorg. Nucl. 
Chem., 27, 129 (1-965); C Kutal and A. W. Adamson, J. Amer. 
Chem. Soc, 93, 558l (1971). 

11. A. C. Sarma, A. Fenertz, and S. T. Spees, J. Phys. Chem., 74 , 
4598 (1970); H. E. Spencer, J. Phys. Chem., 73, 23l6 (1969JT 

12. W. W. Wendlandt and J. H. Woodlock, J. Inorg. Nucl. Chem., 27, 
259 (1965); D. A. Johnson and J. E. Martin, Inorg. Chem., 8TT 
2509 (1969). 

13. L. V. Interrante, K. W. Browall, and F. P. Purdy, Inorg. Chem., 
13, 1158, 1162 (1974). 



-30- 



TRANSITION METAL CATALYZED REARRANGEMENTS 
POLYCYCLIC HYDROCARBONS 



OF 



Gordon F. Stuntz 



March 30, 1976 



In 1969, Woodward and Hoffman 1 published a classic review dis- 
cussing the stereochemical pathways of concerted organic reactions. 
They postulated that the symmetry of the highest occupied molecular 
orbital of a molecule will be preserved throughout any concerted 
transformation. Thus, if the anti-bonding orbitals of the product 
are the only orbitals having the same symmetry as bonding orbitals 
in the reactants, a transition state of very high energy is likely 
to exist and the reaction is said to be "symmetry-forbidden." This 
simple application of basic molecular orbital theory to reaction 
chemistry has proven to be very accurate in predicting whether or 
not a given concerted reaction is likely to occur. 

Recently, a large number of polycyclic hydrocarbons have been 
prepared. Although these molecules are often highly strained, they 
are quite stable because available rearrangement pathways are symmetry- 
forbidden. However, in the presence of many transition metal complexes 
strained polycyclic hydrocarbons are rapidly isomerized to thermo- 
dynamically more stable compounds. 2 " 6 

It has been proposed 7 that the primary function of the transition 
metal in these isomerizations is to supply orbitals of the proper sym- 
metry to accept electrons from the reactant and to donate electrons 
from orbitals of the proper symmetry to the product, thus allowing the 
concerted rearrangement to occur. However, recent work has shown that, 
depending on the catalyst, a variety of non-concerted mechanisms for 
these rearrangements may exist. 



+ 



8 



In the presence of Ag ' cubane(l) isomerizes to cuneane(2). 
Similarly, bishomocubane (3) isomerizes to snoutane(^) . Detailed 







(1) 




(2) 




& 



CO 




O 



^ 



(5) 



kinetic and mechanistic studies 9-11 of the isomerizations of substi- 
tuted derivatives of the bishomocubane and homocubane systems indicate 
that the initial step in these rearrangements is electrophilic attack 
by Ag on an edge of the molecule. This gives rise to a delocalized 
cyclopropylcarbinyl cation which releases Ag giving the final product. 



-31- 



When PdCl 2 is used as a catalyst, the same products are obtained 
as in the presence of Ag . 8 However, with complexes of the type 
PdR 2 Cl 2 , where R is a ligand with weak o donor and strong tt accentor 
capabilities, a mixture of snoutane(4) and tricyclodecadiene(5) is 
obtained from bi shorn ocubane. 1 2 ~ 14 When R is a ligand with a strong a 
donor and strong tt acceptor capabilities, isomerization to the diene 
is the favored process. The amount of diene produced correlates well 
with the ability of the palladium complex to undergo oxidative ad- 
dition. 15 Furthermore, in the presence of complexes of the type 
Rh( diene )ci/ 2 cubane (l) is quantitatively converted into tricylco- 
octadiene(6) # 16 When Rh(co) 2 Cl/ 2 is used, an acylrhodium complex can 
be isolated which must arise from the oxidative addition of a carbon- 
carbon a bond, followed by CO insertion. These results imply that 
the isomerizations can proceed through oxidative addition of a strained 
carbon- carbon bond followed by rearrangement and reductive elimination 
to give the final product. 

Finally, it has been shown that for some Pdll complexes another 
mechanistic pathway may be operative. Cis-bicyclo[6. 1. 0] non-4-ene(7) 
is isomerized to 1, 5-cyclononadiene(8) by PdCl 2 (PhCN) 2 . ^ This pro- 
cess has been shown to proceed through chloropalladation of the cyclo- 
propane ring, followed by 1,2-hydrogen migration and elimination of 
PdCl to give the nonadiene product. It remains to be seen if this 
type of rearrangement pathway is also applicable to the isomerizations 
of the cubane derivatives. 





(7) (8) 

References 

1. R. B. Woodward and R. Hoffman, Angew. Chem. , Int. Ed. Engl., 8, 
781 (1969). 

2. H. Hogeveen and H. C. Volger, J. Amer. Chem. Soc, 89, 2486 
(1967). 

3. L. Cassar and J. Halpern, Chem. Comm. , 1082 (1970). 

4. H. Hogeveen and H. C. Volger, Chem. Comm., 1133 (1967). 

5. W. G. Dauben, M. G. Buzzolini, C. H. Schallhorn and 
D. L. Whalen, Tett. Lett., 787 (1970). 

6. L. A. Paquette and J. C. Stowell, J. Amer. Chem. Soc, 92 , 
2584 (1970). 

7. F. D. Mango, Advan. Catal., 20, 291 (1969). 

8. L. Cassar, P. E. Eaton and J. Halpern, J. Amer. Chem. Soc, 92 , 
6366 (1970). 

9. L. A. Paquette, R. S. Beckley and W. B. Farnham, J. Amer. 
Chem. Soc, 97, 1089 (1975). 



-32- 



10. L. A. Paquette, J. S. Ward, R. A. Boggs and W. B. Farnham, 
J. Amer. Chem. Soc, 97, 1101 (1975). 

11. L. A. Paquette and R. S. Beckley, J. Amer. Chem. Soc, 97, 
1084 (1975). 

12. ¥. G. Dauben and A. J. Kielbania, J. Amer. Chem. Soc, 93 , 
7345 (1971). 

13. L. A. Paquette, R. A. Boggs, ¥. B. Farnham and R. S. Beckley, 
J. Amer. Chem. Soc, 97, 1112 (1975). 

14. L. A. Paquette. R. A. Boggs and J. S. Ward, J. Amer. Chem. Soc, 
97, 1118 (1975). 

15. J. P. Collman, Ace Chem. Res., 1, 136 (1968). 

16. L. Cassar, P. E. Eaton and J. Halpern, J. Amer. Chem. Soc, 
92, 3515 (1970). 

17. G. Albelo and M. F. Rettig, J. Organomet. Chem., 42, 183 (1972). 



-33- 

SOME ASPECTS OF THE OLEFIN METATHESIS REACTION 

William Willis April 6, 1976 

The metal catalyzed olefin metathesis reaction, first reported by 
Banks and Bailey 1 in 1964, can be formally represented as follows: 

R X HC CHR 2 R 1 HC=CHR 2 

R*HC CHR 2 R^C^CHR 2 

Other workers, employing 2 H and 11t C labeling, 3 ' ** confirmed the complete 
breaking of olefinic bonds, and the formation of new olefinic bonds, 
as shown above. This reaction proved to be of industrial importance 5 
(e.g., Phillips Triolefin process for propylene matathesis) and of 
theoretical interest. This discussion will concern the nature of 
metathesis catalysts, the kinds of products obtained, and several 
proposed intermediates involved in the reaction. 

The most effective heterogeneous catalysts are generally oxides or 
carbonyls of W or Mo, usually supported on refractory materials such as 
Si0 2 or AI2O3. Other metal compounds, e.g., Re 2 7 can act as catalysts, 
but are much less effective. The function of the promoter (transition 
metal compound) and support are not clearly divided; changing the 
support will often alter the effectiveness of the catalyst. Reactions 
are usually run at elevated temperature (150° - 500°C) 5 . 

Most homogeneous catalysts consist of a transition metal compound 
and a non-transition metal co-catalyst, often aluminum alkyl halides, 
lithium alkyls, and Grignard reagents. Halides and carbonyls of W 
and nitrosyls of Mo are most effective transition metal constituents. 
Some systems employ an alcohol as a third component of the catalytic 
mixture. 5 It is not clear whether or not the so-called homogeneous 
systems are actually homogeneous. The WCl 6 /EtAlCl 2 /EtOH mixture has 
been debated in this regard. 6-8 Reactions are usually run at room 
temperature, and may or may not use diluents (e.g., alkanes or 
aromatic compounds). 

Most olefin metathesis systems have several notable features: (1) 
Reactions are athermal, i.e., AH~0. (2) Equilibrium is readily 
achieved by using either starting materials or products, i.e., the 
reaction is reversible. (3) The kinds of products obtainable can be 
predicted by random pairing of all available alkylidene groups. 
(4) The concentrations of reactants and products appear to be deter- 
mined by entropic and stereochemical factors. 2 ' 5 Side reactions, 
such as double bond isomerization, also occur. The multi-component 
nature of the catalysts, the complexity of product mixtures, and the 
failure to identify the catalytically active species have hampered 
progress in determining the reaction mechanism. In some cases, the 
concentration of catalyst required to rapidly reach equilibrium may 
be very small, indicating a very labile intermediate. 

While detailed mechanistic information is lacking, several inter- 
mediates have been proposed to adduce the nature and distribution of 
products. The first mechanism suggested the initial coordination of 
two olefins to the metal center, followed by a concerted [2+2] cyclo- 
addition reaction for the formation of a "quasicyclobutane" inter- 
mediate : 



-34- 
I^HC CHR 2 R^C CHR 2 R 1 HC=CHR 2 



->■ 



->- 



R X HC CHR 2 R 2 HC CHR 2 R^CsrCHR 2 

The uncatalyzed reaction is thermally forbidden. Several articles have 
appeared concerning this problem. Each invokes an electron transfer 
through available d orbitals to remove the forbiddenness of the 
reaction. 10 A more detailed theoretical approach has raised questions 
concerning the allowedness of quasicyclobutane formation in the case 
of d 2 10 metal systems. Another problem is the lack of cyclobutane 
formation in all catalytic systems 'reported to date, and the nonreac- 
tivity of free cyclobutanes with catalyst systems. 11 

Later workers proposed an intermediate containing no antisymmetric 
ligand orbitals, thus circumventing symmetry restrictions. 12 In their 
model, the initially coordinated olefins split into four separate 
carbene ligands. The carbene ligands then join to form new olefins. 

R X HC CHR 2 R^C^. ^CHR 2 R 1 HC=y=CHR 2 

i H- M_ " 2 ^ i ^K 2 : i JL 2 

R X HC CHR 2 R^C^ ^CHR 2 R 1 HC= J =CHR 2 

It was later recognized that catalysts derived from W(CO) 6 could have 
only two CO ligands coordinated to the metal under these conditions, 
due to coordinate saturation of the metal - 1 3 Experimental evidence 
does not conclusively rule out this intermediate, however. 

A third mechanism requires the formation of a 5-membered metallo- 
cycle. 11 * The authors do not propose a pathway for step b. 

RHC CHR RHC CH 2 RHC=f=CH 2 

|| — M — 1| RHC CHR / \ + ip 

H 2 C CH 2 «- / \ ^RHC V .CH 2 *" RHC=CH 2 



H 2 C CH 

M 



k M' 



This mechanism was based on the reaction of 1 ,4-dilithiobutane with 
WC1 6 , which resulted in the formation of ethylene. 

A forth mechanism has been proposed in which a carbene and a 
coordinated olefin form a 4-membered metallocycle : 

R 1 HC=f=CHR 1 X RHC CHR 1 X RHC : RHC CHR 1 

5* | | 5± || CHR 1 ^± (I + II 
M=CHR 2 M CHR 2 M 1| M CHR 2 



CHR 



2 



Attack by R 2 HC=CHR 2 on the last species results in the formation of 
more mixed alkylidene product R 1 HC=CHR 2 . Several papers have appeared 
giving evidence for the intermediacy of coordinated carbenes. 15 
This has been demonstrated in particular by the isolation of catalytically 
active carbene intermediates which result from the reaction of L(PH3P) 2 
RhCl and 

R 1 R 2 



f^ N \ X N ^| 



-35- 

CONCLUSION 

The mechanism of the olefin metathesis reaction remains unsolved, 
and will require further investigation of simpler catalytic systems 
from which active intermediates can be isolated, or otherwise identified 
Of the several proposed pathways, the 4-membered metallocycle mechanism 
appears to present the fewest difficulties, and should provide 
additional direction for future work. 

References 

1. R. L. Banks and G. C. Bailey, I. & E. C. Prod. Res. Dev. , 3, 170 
(1964). 

2. N. Calderon, E. A. Ofstead, J. P. Ward, W. A. Judy and K. W. Scott, 
J. Amer. Chem. Soc, 90, 4133 (1968). 

3. F. L. Woody, M. J. Lewis, G. B. Wills, J. Catal., 14, 389 (1969). 

4. J. C. Mol, J. A. Moulijn, and C. Boelhouwer, Chem. Comm. , 1968 , 633 

5. R. J. Haines, G. J. Leigh, Chem. Soc. Rev., 4, 155 (1975). 

6. E. L. Muetterties, M. A. Busch, J. Chem. Soc., Chem. Commun. , 

1974 , 751. 

7- R. Wolovsky, Z. Nir, J. Chem. Soc., Chem. Commun., 1975, 302. 

8. M. T. Mocella, M. A. Busch, and E. L. Muetterties, J. Amer. Chem. 
Soc. , 98, 1283 (1976). 

9. C. P. C. Bradshaw, E. J. Howman, and L. Turner, J. Cata., 7_, 269 

(1967). 

10. F. D. Mango, J. H. Schachtschneider , J. Amer. Chem. Soc, 93 , 

1123 (1971). 

11. G. L. Caldow, R. A. MacGregor, J. Chem. Soc, A, 1971, 1654. 

12. G. S. Lewandos, R. Pettit, Tetrahedron Lett., 1971 , 789. 

13. G. S. Lewandos, R. Pettit, J. Amer. Chem. Soc, 93, 7087 (1971). 

14. R. H. Grubbs, T. K. Brunck, J. Amer. Chem. Soc, 94, 2538 (1972). 

15. D. J. Cardin, M. J. Doyle, and M. F. Lappert , J. Chem. Soc., Chem. 
Commun. , 1972 , 927. 

16. M. F. Farona, W. S. Greenlee, J. Chem. Soc., Chem. Commun. 

1975 , 759. 

17. C. P. Casey, T. J. Burkhardt , J. Amer. Chem. Soc, 96, 7808 (1974). 

18. R. H. Grubbs, P. L. Burk, D. D. Carr, J. Amer. Chem. Soc, 97, 
3265 (1975). 



-36- 

ONE DIMENSIONAL METALLIC CONDUCTORS DERIVED FROM THIRD ROW 

d 8 TRANSITION METAL CCMPLEXES 

Anton Elamma April 20, 1976 

In the solid state, the partially oxidized complexes of Pt(ll) 
and Ir(l) are one-dimensional (l-D) electronic conductors. 1 > 2 * 3 > 4 
These compounds form parallel linear chains 5 ' 6 ' 7 and have short metal- 
metal bonds like those in pure metals. Unoxidized complexes such as 
K 2 Pt(CN) 4 also form linear chains 8 in which extensive Pt-Pt inter- 
actions exist but are not electronic conductors. It has been suggested 9 
that the interactions of primary concern involve the d 2 orbitals on 
the metal centers which overlap in the solid state to form a valence 
band of appreciable width. The energy level of d 2 in [Pt(CN) 4 ]~ is 
still debatable; 10 nevertheless, the upper levels of the valence band 
are antibonding. The removal of electrons from this part of the band 
by partial oxidation strengthens the bonding in the chain and forms a 
partially filled band which by definition is metallic. 

The partially oxidized phase K 2 [Pt(CN) 4 ]Br Q . 3O .nH 2 0(KCP) has opti- 
cal properties similar to metals 11 and a room temperature conductivity 2 
a = 300 ohm" 1 cm" 1 . This simple picture does not account for the smooth 
decline of the conductivity with decreasing temperature leading to an 
insulator at 20°K, a = 10" 12 ohm" 1 cm" 1 . 

Peierls 12 has shown that, in the absence of electron- lattice vi- 
bration interactions, a l-D metal is an unstable structure. It under- 
goes a spontaneous lattice distortion which by the creation of an 
energy gap at the Fermi level splits the metallic band into filled and 
empty bands. Consequently, the energy of the system is lowered, and a 
l-D metal becomes an intrinsic semiconductor or an insulator. The 
Peierls' distortion is described using the arguments of Tight -Binding- 
Band-Theory, which is the LCAO approach for crystalline solids. 13 

' The phonon spectrum 14 of KCP using neutron scattering techniques 
shows a Kohn 15 anomaly. A Kohn anomaly arises from the interaction of 
lattice vibrations (phonons) with electrons. The phonons are quantized 
and each has an assigned wave vector q = 2tt/\ (X is the wavelength of 
the phonon). The lattice favors the phonons which enhance the mani- 
festation of a Peierls' distortion since this lowers the total kinetic 
energy of the electrons. When the harmonic restoring forces in the 
lattice no longer have sufficient thermal energy to return the atoms 
or molecules to the undistorted equilibrium configuration, the energy 
associated with the Peierls' effect dominates and a metal- insulator 
transition takes place. 

The features necessary for l-D conducting systems are discussed. 
Synthetic routes 16 * 1T , 1Q and conductivity-measurement techniques 19 
are described for the partially oxidized "phases. 



-37- 

References 

Complete reviews are found in references 1 and 2. 

1. H. R. Zeller, Festkoerperprobleme, 13, 31 (1973). 

2. J. S. Miller and A. J. Epstein, Prog. Inorg. Chem., 20, 1 (1976). 

3. D. Kuse and H. R. Zeller, Phys. Rev. Lett., 27, lo6o~Tl97l). 

4. H. R. Zeller and A. Beck, J. Chem. Phys. Solids, 35, 77 (1974). 

5. J. M. Williams, M. Iwata, S. W. Peterson, K. Leslie, 

H. J. Guggenheim, Phys. Rev. Lett., 34, 1653 (1975). 

6. J. M. Williams, J. L. Petersen, H. M. Gerdes, and S. W. Peterson, 

Phys. Rev. Lett., 33, 1079 (197*0. 

7. K. Krogmann, W. Binder and H. D. Hausen, Angew. Chem. Int. Ed. 

Engl., 7i 812 (1968); Angew. Chem., 80, 844 (1968). 

8. D. M. Washecheck, S. W. Peterson, A. H. Reis, Jr., and 

J. M. Williams, Inorg. Chem., JL5, 74 (1976). 

9. K. Krogmann, Angew. Chem. Int. Ed. Engl., 8, 35 (1969). 

10. L. V. Interrante and R. P. Messmer, in Extended Interactions 

between Metal Ions (Leonard V. Interrante, Ed. ) , Chapt . 27, 
American Chemical Society, Washington, 1974. 

11. J. Bernasconi, P. Bruesch, D. Kuse, and H. R. Zeller, 

J. Phys. Chem. Solids, _35, 145 (1974). 

12. R. E. Peierls in Quantum Theory of Solids (Oxford University Press, 

London, 1955), p. 10b*. 

13. B. C. Gerstein, J. Chem. Educ, 50, 3l6 (1973). 

14. B. Renker, H. Rietschel, L. Pintschovius, W. Glaser, P. Bruesch, 

D. Kuse, and M. J. Rice, Phys. Rev. Lett., 30, 1144 (1973). 

15. W. Kohn, Phys. Rev. Lett., 2, 393 (1959). 

16. W. Knop, Justus Liebeg's Ann. Chem., 43, 111 (1342). 

17. K. Krogmann and H. D. Hausen, Z. Anorg. Allg. Chem., 358, 67 (1968) 

18. A. P. Ginsberg, J. W. Koepke, J. J. Hauser, K. W. West, 

F. J. DI Salvo, C. R. Sprinkle, and R. L. Cohen, Inorg. Chem., 
15, 514 (1976). 

19. H. C. Montgomery, J. Appl. Phys., 42, 2671 (l97l). 



-38- 



THERMODYNAMIC AND SPECTROSCOPIC STUDIES OF REACTIVITY AND 

BONDING IN TRANSITION METAL LEWIS ACID-BASE SYSTEMS 

Marinda P. Li (Final Seminar) April 22, 197 6 

Introduction 

Spurred by the growing concern with homogeneous catalysis, 
the reactivity and bonding of organometallic complexes of Group VIII 
metals having a d 8 configuration have been the subject of widespread 
attention 1 * 2 among inorganic chemists during this past decade. In 
many of these catalytic systems, it is felt that a critical step in 
the mechanism involves coordination of the substrate to the metal, 
which can be viewed as a Lewis acid-base interaction. Thus, our 
initial approach to understanding the behavior of such heavy metal 
systems in catalysis was to obtain quantitative knowledge of the 
Lewis acidity of diineric rhodium (i) and palladium (il) systems in 
their behavior towards a variety of Lewis bases. The elucidation of 
factors affecting the strength of interaction may aid in the future 
design of catalytic systems. As yet, the quantitative effects of 
introducing a second metal complex into the coordination sphere of 
a first metal center remain relatively unexplored. Only when con- 
cepts of reaction mechanisms are developed to a point where one can 
predict with confidence new syntheses which can be put into experi- 
mental practice, will the field of homogeneous catalysis by organo- 
metallic compounds have acquired a more rational then empirical basis. 

Comparison of Acid-Base Parameters and Bridge Cleavage Energies for 
Heavy Metal Dimers 

Analogous bridge- splitting reactions have been reported for a 
variety of Lewis bases reacting with the chloro-bridged dimers 
[Rh(l,5-cyclooctadiene)ci] 2 , 3 [Rh(co) 2 Cl] 2 , 4 and Pd(TT-methylallyl)- 
Cl] 2 . 5 It is of interest to examine the effects of changing the metal 
and its oxidation state as well as the terminal substituents on the 
calorimetrically measured enthalpies of adduct formation for these 
heavy metal dimers. A quantitative comparison of acid-base parameters 
and bridge cleavage energies for these three systems is made. 

Development and Tests of a Proposed Acid-Base Model for Mixed Metal 
Dimer Formation 

n 

One can show that when the A/E. ratios of a series of acids 
are similar, the E and C equation 6 can be converted from a double- 
scale enthalpy equation to a single- scale enthalpy relation. 

-AH = E A P B (1) 

where P B = E B +[^] C^ (2) 




-39- 

Knowing the energies required to dissociate the symmetrical dimers 
and the E» and P_ parameters describing the acidity and basicity 
of the monomeric acids and chloride bridges respectively, one is 
in a position to predict the net enthalpy expected for the reaction 
of mixing two symmetrical dimers to form an unsymmetrical mixed 
dimer species. 1 H nmr, 13 C{ 1 H} nmr, and infrared spectroscopy are 
employed to investigate the formation of certain mixed dimers and 
to test our predictions for mixed dirner formation. 

From this research, a general guideline for the design and 
synthesis of mixed dimers has emerged. In general, for any case 
where the C/E ratios for the two metal centers is similar and the 
P R value of the bridging ligands comparable, there will be no net 
enthalpic driving force for forming the mixed dimer regardless of 
the acid strength of the two metal centers or the relative values 
of C-o and E R which lead to the same P„. in order to maximize the 
enthalpic driving force for forming any mixed dimer system, it is 
necessary to increase the P B parameter of the bridging atom(s) on 
the less acidic center and fiecrease the P_ parameter of the bridging 
ligand on the more acidic center. Once again, the obvious value of 
quantitative assessments of acid and base parameters for various 
systems is apparent. 

References 

1. J. P. Collman and W. R. Roper, Adv. Organomet. Chem., 7, 53 (1968). 

2. R. E. Harmon et. al. , Chem. Rev., 73, 21 (1973). 

3. J. Chatt and L. M. Venanzi, J. Chem. Soc, 4735 (1957). 

4. D. N. Lawson and G. Wilkinson, J. Chem. Soc, 1900 (1965). 

5. K. Vrieze et. al. , Rec. Trav. Chim. , 85, 1077 (1966). 

6. R. S. Drago, Structure and Bonding, 15, 73 (1973). 



-40- 

ADDITION OF NUCLEOPHILES TO COORDINATED OLEFINS 

Dennis Sepelak . April 27, 1976 

Although addition of electrophiles to olefins occurs easily, 
addition of nucleophiles only occurs when the double bond is acti- 
vated by the presence of electron withdrawing substituents. Creation 
of a partial positive charge by inductive or resonance effects is 
necessary to allow attack by the nucleophile on the electron rich al- 
kene. 1 Complexation by a transition metal alters the bonding in an 
olefin and can activate the double bond toward attack. A low valent 
metal can increase susceptibility toward electrophilic additions, 
whereas an olef in-metal complex with the metal in a higher oxidation 
state (e.g., Pd(ll)) facilitates attack on the olefin by nucleophiles. 
This is a result of partial withdrawal of electron density from the 
double bond by the positively charged metal. 3 ' 4 



\/ 
c 

m— II mc' 



\/ 

.-c \ /. 

c 

Pd 5 ^<- 



« + 



c c c 6 + 

/ \ / \ / \ 

Although stable metal- olefin tt complexes are known for most of 
the transition metals from the chromium subgroup to the copper sub- 
group, 5 most of the work involving addition of nucleophiles to 
coordinated olefins has been done with Pd(ll), Pt(ll), and Fe(ll) 
(as the [(Ti 5 -C 5 H 5 )Fe(CO) 2 alkene] complexes). Palladium(ll) olefin 
complexes undergo the greatest variety of nucleophllic additions, 
both stoichiometric and catalytic. g > t » 8 j 9 

Olefin complexes of the (n 5 -C 5 H 5 )Fe(C0) 2 ion can be prepared in 
a number of ways, including protonation of the n 1 -allyl compounds and 
B-hydride abstraction from (Ti 5 -C 5 H5)Fe(C0) 2 alkyls. However, for syn- 
thetic purposes, olefin exchange with the (rr'-CsHs )Fe(C0) 2 (isobutylene) 
cation and the reaction of Na(Tn 5 -C 5 H 5 )Fe(co) 2 with epoxides are probably 
most useful. 10 The latter reaction occurs with retention of configur- 
ation. 11 These complexes undergo addition by alcohols, amines, enamines, 
cyanide and phosphines as well as anionic nucleophiles such as enolate 
anions. The reactions are frequently regiospecific. x 2 » 13 Reactions 
of similar complexes of other metals have been studied. 14 




(T] 5 -C 5 H 5 )Fe( 00) p*-|— + HOCHg *(Ti 5 -C 5 H5)Fe(C0) 2 - 



■OCH: 



CH- 



Olefin complexes of palladium ( II ) can be formed by simply adding 
olefins to solutions of Pd(ll) salts, such as [PdCl 4 ]" 2 . 5 It was 
known for quite some time that palladium-ethylene complexes undergo 
stoichiometric reaction with water to yield acetaldehyde and palladium 
metal. 15 However, it was not until J. Schmidt and coworkers at 



-41- 



Wacker Chemie 16 observed that the palladium could be oxidized back to 
Pd(ll) by such agents as Cu(ll) (which could in turn be oxidized by 
atmospheric oxygen) that extensive work was done in this area. 

Elucidating the kinetics and mechanism of the olefin oxidation 
reaction was the object of much work by a number of workers. 17 > 1B > 19 
The evidence suggests a mechanism in which there is nucleophilic 
attack by a coordinated hydroxo group cis to the olefin. 




H 
HO-CH CH— PdCl. 



Since the discovery of the Wacker process, addition of a variety 



of nucleophiles has been observed, including alcohols, 20 amines, 
enolate anions, carboxylic acids, 22 arenes, 23 and chloride ion. 
actions also occur with dienes, allenes, and butadienes. 



21 



Re- 



The stereochemistry of the addition to dienes and bidentate ole- 
fins was found to be trans, indicating addition of noncoordinating 
nucleophiles. 24 




weak base 
CK3QH 





27 
28 



M«Pt 
M=Pd 



Trans addition was also found to be the case in reactions of 
monoolefins with alcohols, 25 acetate, 26 and amines. 27 Additions in 
which the nucleophile is coordinated to the metal prior to reaction 
appear to occur through anti-Markownikoff cis addition and elimination, 
whereas addition of none coordinated nucleophiles is Markownikoff with 
trans addition and elimination. 25 

References: 



1. S. Patai, The Chemistry of Alkenes , pg. 469, Interscience, 1964. 

2. P. M. Maitlis, The Organic Chemistry of Palladium , Vol. I and II. 

Academic Press, 1971. 

3. M. K. Chisholm and H. C. Clark, Accounts Chem. Res., 6, 202 (1973) 

4. E. 0. Graves, C. J. L. Lock and P. M. Maitlis, Can. J. Chem., 

46, 3879 (1968). 

5. M. Herberhold, Metal tt Complexes , Vol. II, Part I; Elsevier, 1972. 

6. P. M. Henry, Adv. Organometal lie Chemistry, Vol. 13, pg. 363, 

Academic Press, 1975. 

7. Symposium on Homogeneous Catalytic Reactions Involving Palladium, 

Am. Chem. Soc. Petroleum Preprints, _l4, No. 2, B- 1-172, 1969. 



-42- 

8. J. Tsuji, Accounts Chem. Res., 2, 122 (196S) ; 6, 8 (1973). 

9. J. Tsuji, Adv. Organic Chem., Vol. 6, pg. 109, Interscience, 1969. 

10. M. Rosenblum, Accounts Chem. Res., 7, 122 (1974). 

11. W. P. Giering, M. Rosenblum and J. Tancrede, J. Am. Chem. Soc, 

94, 7171 (1972). 

12. L. Busetto, A. Pallazzi, R. Ros and U. Belluco, 
J. Organomet. Chem., 25, 207 (1970). 

13. A. Rosen, M. Rosenblum and J. Tancrede, J. Am. Chem. Soc, 

95, 3062 (1973). 

14. ¥7 H. Knoth, Inorg. Chem., 14, 1566 (1975). 

15. F. C. Phillips, Am. Chem. J.,.l6, 255 (l894). 

16. J. Schmidt, et. al., Angew. Chem., 74, 93 (1962). 

17. R. Jira and W. Freiesleben, Organometallic Reactions, Vol. 3, 
pg. 1, Wiley, 1972. 

18. R. F. Heck, Forschr. Chem. Forsch. , 16, 221 (1971). 

19. P. M. Henry, J. Am. Chem. Soc, 86, 3246 (1964). 

20. A. D. Ketley and L. P. Fisher, J. Organomet. Chem., 13, 243 (1968). 

21. H. Hirai, H. Sawai and S. Makishima, Bull. Chem. Soc. Japan, 45 , 
1148 (1970). 

22. W. Kitching, Z. Rappoport, S. Winstein and W. G. Young, 
J. Am. Chem. Soc, 88, 205^ (1966). 

23. S. Danno, I. Moritani and Y. Fu.iiwara, Tetrahedron, 25 , 
4809 (1969). 

24. J. K. Stille and R. A. Morgan, J. Am. Chem. Soc, 88, 
5135 (1966). 

25. D. E. James, L. F. Hines and J. K. Stille, J. Am. Chem. Soc, 
98, 1806 (1976). 

26. P. M. Henry and G. A. Ward, J. Am. Chem. Soc, 93, 1494 (1971). 

27. B. Akermark, J. Bachvalll and K. Surala-Hansen, Tetrah. Letters, 
1363 (1974). 



■43- 



:n organic chemistry in the gas phase as revealed by ion 
cyclotron resonance spectroscopy 1 
Patrick Cannady May 4, 1976 

m (. JCTION 

The motion of an ion in a uniform magnetic field is restricted 
t circular orbit in a plane perpendicular to the direction of the 
PI Ld Electromagnetic energy of a frequency equal to the cyclotron 
- requency, t» , 



CD. 



qH 

nc 



-rd absorbed by 



.7 s ,cn ions, 
nargi al oscillator detector. With 

>nstant, sweeping the magnetic field 
is is known as the single resonance 

•r- sing energies (^70ev) yields the 



and the absorption of power detected by 
the observing frequency held 
yields a spectrum linear in mass 
spectrum, and at high electron 
mass spectrum of the sample. 



By varying the electron energy, appearance potentials can be de- 
termined and reactions of primary and secondary ions with the parent 
neutral species at various pressures can be investigated. The tech- 
lique of double resonance as applied to ion cyclotron resonance 
(icr) 2 allows for the unequivocal identification of reaction pathways 
for these ion-molecule reactions. In this technique, a second radio- 

iuency field is swept as the first radiofrequency field and the mag- 
netic field are held at resonance conditions for the product ion under 
iy. As the second RF field comes into resonance with the precursor 
.ons, the energy absorbed will cause a change in the reaction rate, 
iltering the rate of product ion observation, thus identifying the pre- 
cursors of the product ion. 

Trapped ion techniques, 4 in which ions produced by electron impact 
.d in the source region by an electromagnetic force field for a 
eriod of time before detection, allow for direct observation 
the time dependence of product distributions and abundances. The 
-;_ action of kinetic parameters is thus greatly simplified. 

LI CATION 



Ion-molecule reactions in the gas phase are rather conveniently 
.nvestigated by icr methods. Gas phase Lewis acidities and basicities 
>f a number of inorganic hydrides as well as those of small organic 
lolecules have been and continue to be investigated. Gas phase basici- 
-ies are characterized by the proton affinity (PA), which is defined as 
:he enthalpy change of the reaction 

MH + -> M + H + 

and by the hydrogen affinity (HA), or, as it is often referred to, the 
hemolytic bond dissociation energy, D(M -H), which is the enthalpy 



44- 



change of the reaction 



,+ 



+ 



MH -> M + H 



These trio quantities are related by the ionization potential of M, 
by thb equation 

PA(M) - D(M + -H) = IP(H) - IP(M) 

Within a homologous series, the hydrogen affinity is usually 
found to be a constant, resulting in a direct parallel between the 
proton affinity and the ionization potential of the base. Thus, the 
order of basicities in such a series follows the order of the induc- 
tive electron donating abilities of the substituents. For example, 
the proton affinities of the methyl amines increases in the order 
PA' N Hh) < PA(HN(<Jii 3 )j: < PA(n(CH 3 )) in the gas phase. In contrast, 
ition basicities yield an inverted order, PA(H 2 NCH 3 ) < PH(HN(CH 3 ) 2 
m!m'(CH 3 )3. The inverted order is due to complicating solvent ef- 
wJv/S, which are totally lacking in the gas phase. 5 

Acidities can be measured in terms of other ion affinities than 
proton affinities. Inorganic anions such as F have been used to 
determine Lewis acidities in the gas phase of some inorganic fluorides. 6 
Relative F~ ion affinities (IA = AH for the reaction: 
AX - / \ ■* A/ \ + X"/ \) are found to increase in the order SF 4 , 
SF 5 'v S0 2 , ^SF 3 < S?F 4 < BF 3 < PF 5 < BC1 3 < AsF 5 , which is in agreement 
with earlier solution work done in CH 2 C1 2 . 7 

Proton affinities and hydrogen affinities have been measured in 
the gas phase for such organometallic species as Fe(co) 5 , Fe(CsH 5 ) 2 , 
and Ni(C 5 H5) 2 and are summarized below: B > 9 > 10 



Fe(C0) 5 

Fe(C 5 H 5 ) 

Ni(C 5 H 5 ) 



PA 

198 kcal/mol 
207 kcal/mol 
218.9 kcal/mol 



HA 

74+3 kcal/mol 
56.6+6 kcal/mol 
48.3+3 kcal/mol 



The positive ion chemistry of Fe(co) 5 9 > 1 °> 11 is quite rich, with 
evidence found for the formation of polynuclear Fe x (C0) ions, where 
x ranges from 2 to 4, and n from 3 to 12. Higher polynuclear species 

e postulated, but as yet an upper limit of m/e=6oo cannot be ex- 
ceeded with present experimental configurations. Ligand displacement 
reactions between Fe(co) (n=l-5) have been investigated for a series 
of ligands. Fe(co) 5 itself is found to be rather inert to substi- 
tution, whereas Fe(C0) 3 ' is most active. The gas phase basicities of 
the Fe(co) 5 , Fe(c 5 H 5 ) 2 , and Ni(C^H 5 ) 2 species are rather large 
(PA(NH 3 ) = 201 kcal/mol). Fe(co) 5 and Fe(c 5 H 5 ) 2 are known to protonate 
only in highly acidic solutions, with proton-metal bonds being formed. 
Studies 8 have been directed towards discerning the site of protonation 
of Fe(c 5 H 5 ) 2 and Ni(C 5 H 5 ) 2 in the gas phase, but the results are as 
yet inconclusive. The negative ion chemistry of Ni(C0) 4 and Cr(C0) 6 
has also been investigated and is dominated by the formation of 
binuclear metal complexes. 



-45- 

CONCLUSION 

TCR spectroscopy has tremendous potential as a tool for the 
investigation of certain intrinsic aspects of organometallic chemistry 
in the gas phase, in the absence of complicating solvent phenomenae. 
Areas of particular interest may be: (l) polynuclear metal cluster 
• or^nation, (2) ligand substitution processes, (3) relative ligand 
binding energy determinations, (4) transition metal % basicities, 
(5) processes involving both electrophilic and nucleophilic attack 
on neutral metal complexes, (6) unusual a- and n-bonded organometallic 
complexes, and (7) photochemistry of gaseous organometallic compounds. 

REFERENCES 

1. A general review of icr is found in J. L. Beauchamp, 
&nnu. Rev. Phys. Chem., _22, 527 (1971). 

2 L. R. Anders, J. L. Beauchamp, R. C. Dunbar, and 

J. D. Baldeschwieler, J. Chem. Phys., 45, 1062 (1966). 

3. B. S. Freiser, T. B. McMahon and J. L. Beauchamp, 
Int. J. Mass. Spectrom. Ion Phys., 12, 249 (1973). 

4. T. B. McMahon and J. L. Beauchamp, Rev. Sci. Instrum., 43, 
509 (1972). 

5. E. M. Arnett, Ace. Chem. Res., 6, 4o4 (1973). 

6. J. C. Haartz and D. H. Daniel, J. Amer. Chem. Soc, 95* 
8562 (1973). ~" 

7. S. Brownstein, Can. J. Chem., 47, 605 (1969). 

8. Reed R. Corderman and J. L. Beauchamp, Inorg. Chem., 15 , 
665 (1976). 

9. M. S. Foster and J. L. Beauchamp, J. Amer. Chem. Soc, 93, 
4924 (1973). 

10. M. S, Foster and J. L. Beauchamp, J. Amer. Chem. Soc, 97, 
4808 (1975). 

li. R. C. Dunbar, J. F. Ennever and J. P. Fackler, Jr., 
Inorg. Chem., 12, 2734 (1973). 



-46- 

SIGMA-PI REARRANGEMENTS IN ORGANOMETALLEC SYSTEMS 
Keith Hodges May 11, 1976 



In 1957 it was first recognized that an organic grouper- bonded 
to a transition metal could rearrange to a n-bonded species. This 
discovery came with the elucidation of the structures of bisbiphenyl 
chromium(o) and benzene binhenyl chromium ( 0) ' (Rein's complexes) and 
the isolation of the intermediate in their preparation, tri-a-phenyl 
chromium tristetrahydrofuranate. 2 Since then these rearrangements have 
been observed for a wide variety of organ ometallic compounds and 
several industrially important processes rely on reactions involving 
q-tt rearrangements; most notably, the Ziegler-Natta process for the 
polymerization of olefins, the Wacker process for the oxidation of 
ethylene to acetaldehyde, and the " oxo" process for the hydroformyl- 
ation of olefins. 

Sigma-pi rearrangements (and the converse, pi-sigma rearrange- 
ments) are intramolecular processes that can be classified in two 
general categories. The first classification involves reactions of 
the organic moiety that induce rearrangement. For example, many a- 
bonded allyl complexes are readily protonated by mineral acids to 
give the corresponding n-nropene cationic complexes. Specific examples 
include: 3,A 



H 



+ 



(Ti 5 -C 5 H5)Pe(C0) 2 (r 1 -C 3 H5) > \ (•p 5 -(: 5 H 5 )Fe(CO) 2 (ti 2 -C 3 H 6 ) ] 



+ 



(CO) 5 Mn('n I -C3H5) 



II 



[(C0) 5 Mn(T 1 2 -C3H 6 )] 



Such reactions are not limited to allyl complexes, however, and have 
been observed for numerous unsaturated organic groups bound to a 
metal, e.g., 5 ' 6 



(Tl 5 -C5H 5 )j''e(C0) 2 (in 1 -Cc 5 Hc3) 



(Tl 5 -C 5 H5)Fe(C0)o(iV-0HoCN) 



II 



H 



[(Ti- > -c: 5 H l 5)Fe(co)p(n 2 -c 5 H 6 )] 



+ 



(ri 5 -C 5 H5)Fe(C0) P ' 

t 

N-OCH2 
/ 

H 

Alternatively, abstraction of a hydride ion from the organic group of 
a metal-alkyl complex can generate a cationic n-alkene complex. 
Interestingly, this reaction is often reversed on the addition of 

u." 

-H' 

(r 1 5 -C 5 H 5 )Fe(CO)pC 2 H ri rt [ (ry'-C 5 H 5 )Fe (CO) 2 {r\ p -C P lU ) ] 

flf 

-H" 
(•n 5 -C 5 H 5 )Mo(C0)3C 2 H 5 * [(ti 5 -C5H5)Mo(C0) 3 (t 1 ? -C 2 H4)] 

:-H~ 



NaBJL 7 > 8 



-4Y- 



The second classification of sigma-pi rearrangements involves the 
addition or abstraction of ligands bound to the metal to induce a co- 
ordination change in the organic group. Pome reactions are reversible, 
and the most numerous examples exist for carbonyl complexes. Fe, 9 
Mn, 10 Mo 11 and W 12 carbonyls with a-bound organic substituents rearrange 
to the corresponding n-cornplex upon loss of CO, e.g., 

-CO 

(CO) 5 Mn-CH 2 CH-C}ICHo =s (co) 4 Mn 

i-CO 




(•n 5 -c 5 H 5 )Mo(co) 



CHpSCHi 



-co 



ICO 



(ti 5 -C 5 H5)Mo(C0) 2 

t 

CH 2 -S-CH 3 



Analogous reactions have been observed In cobalt cyanide complexes. 
The a-butenyl complex obtained by the addition of butadiene to hydrido- 
pentacyanocobaltate(lll) can eliminate cyanide ion to give a n-butenyl 
complex. 13 Similar steps have been proposed in the mechanism of the 
stereoselective hydrogenation of butadiene by pentacyanocobaltate(ll) 
in the presence of II ; 



14 



Attempts have been made to show that addition or elimination of 
ligands and concomitant rearrangements are a function of the rr-bonding 
strength of the ancillary ligands in the complex. For a-bonded com- 
pounds a correlation has been drawn between their stability and the 
possibility of electronic transitions from high energy filled d orbitals 
into the lowest antibonding molecular orbitals corresponding to the 
first step in metal-carbon bond rupture. 12 Increasing the energy of 
this transition (AE) by the addition of ligands which stabilize the 
nonbonding d orbitals should result in a more stable a-complex. In 
rr-complexes, AE is increased by the interaction of the metal d orbitals 
and the antibonding tt~ orbitals of the organic group. Thus the a- state 
will be destabilized with respect to the tt- state on losing a ligand 
and a q-tt rearrangement can occur. The reverse can occur upon addition 
of a strong rr-bonding ligand to a rr-complex. This leads one to predict 
that addition of ligands such as CO, CN~, PR 3 and DMSO to n-complexes 
should result in pi-sigma rearrangements. 



CONCLUSIONS: 



'hi; 



research into catalvtic 



area of study provides a fertile ground for 
systems, the activation of C-H bonds and the 



preparation of inherently unstable organic molecules that are diffi- 
cult or impossible to prepare by conventional routes. 



REFERENCES 



1. 
2. 
3. 

h. 



5. 
6. 



H. H. Zeiss and M. Tsutsui, J. Amer. Chem. Soc 
W. Herzig and H. H. Zeiss, J. Amer. Chem. Soc. 
M. L. H. Green and P. L. I. Nagy, J. Chem. Soc, 
M. L. H. Green, A. G. Massey, J. T. Moelwyn- Hughe 
late) P. L. I. Nagy, J. Organometal. Chem., 



(the 
M. L. 
J. K. 



79, 3062 (1957). 
79, 6561 (1957). 
189 (1963). 
and 
8, 511 (1967) 



H. Green and P. L 
P. Ariyartine and 



I. Nagy, Z. Naturforsch. 
M. L. II. Green, J. Chem. 



18b , 162 (1963). 
»c. , 2976 (1963). 



-*to- 



7. M. L. H. Green and P. L. I. Nagy, J. Organometal. Chem., 1, 
58 (1963). 

8. M. L. H. Green and M. Cousins, J. Chem. Soc, 889 (1963). 

9. S. Savel, R. Ben-Shoshan, and B. Kioson, J. Amer. Chem. Soc. , 
87, 2517 (1965). 

10. W. R. McClellan, H. H. Hoeln, H. N Crips, E. L. Muetterties, 
and B. W. Houk, J. Amer. Chem. Soc., 83, l6oi (l96l). 

11. R. B. King and M. B. Bisnette, Inorg. Chem., k , 486 (1065). 

12. F. A. Cotton and T. J. Marks, J. Amer. Chem. Soc, 91, 1339 (1969) 

13. J. Kwiatek and J. K. Seyler, J. Organometal. Chem., 3, 421 (1965 

14. J. Kwiatek and J. K. Seyler, J. Organometal. Chem., 3, 433 (1965 

15. J. Chatt and B, L. Shaw, J. Chem. Soc, 705 (1959). 



- TABLE OF CONTENTS 

INORGANIC SEMINAR ABSTRACTS 

1976-1977 
Summer Session : Page 

INVESTIGATIONS INTO THE ELECTRONIC STRUCTURE OF DIOXYGEN 1 
COORDINATED TO COBALT (il) COMPLEXES - David J. Kitko 

TEMPERATURE 'DEPENDENT STUDIES OF STRUCTURAL AND 4 

ELECTRONIC EFFECTS IN SINGLE CRYSTALS - Gerald Delker 

THE MECHANISM OF Nl(l) -CATALYZED OLEFIN ISOMERIZATION - 5 

Michael J. D'Aniello, Jr. 

STUDIES OF CYCLOPENTADIENYLBISLIGANDNICKEL(I) COMPLEXES - 7 
Alex N. Williamson 

ELECTRON EXCHANGE INTERACTIONS IN COPPER ( II ) AND 8 

MANGANESE (II) OUTER-SPHERE-ASSOCIATED DIMERSj THE DYNAMICS 
OF ELECTRON TRANSFER - Edward J. Laskowski 

NOVEL STUDIES OF TRANSITION METAL DIMERS - 10 

Robert M. Richman 

ELECTRON EXCHANGE INTERACTIONS IN DIMERIC LINEAR BRIDGED 12 
COMPLEXES OF CU(ll) AND IN DIBRIDGED COMPOUNDS OF Tl(lll) - 
Rudolph G. Jungst 

Fall Session : 

THE IMPORTANCE OF SYMMETRY IN THE UNDERSTANDING OF PHASE 14 
CHANGES IN CERTAIN THREE-DIMENSIONAL ANTIFERROMAGNETS - 
Jeffry A. Kelber 

HETEROGENEOUS CATALYSIS BY ZEOLITES - David J. Blumer 15 

MttSSBAUER SPECTROSCOPY APPLIED TO SUPPORTED METAL CATALYSIS - l8 
Lynn c. Francesconi 

HIGH POTENTIAL IRON SULFUR PROTEIN AND THE THREE 20 

STATE HYPOTHESIS - Carol Iris Ashby 

STRUCTURE OF CYCLOPHOSPHAZENES IN RELATION TO THEIR 23 

METAL COMPLEXES - Gerald V. Rubenacker 

cis-DIAMMINEDICHLOROPLATINUI-i(ll) AND ITS ANALOGS: 26 

THE SEARCH FOR A CURE FOR CANCER - Peter A. Bellus 

STUDIES OF SOME DYNAMIC METAL CARBONYL CLUSTER COMPLEXES - 28 
Steve Richter 

PHOTOCHEMICAL SUBSTITUTION OF MANGANESE DECACARBONYL - 29 

Dennis Kidd 



Spring Semester : 

HIGH PRESSURE EFFECTS ON TRANSITION METAL COMPLEXES - 30 

Virgil L. Payne 

RADICAL CHAIN PATHWAYS IN SUBSTITUTION REACTIONS OF METAL 32 
CARBONYL COMPOUNDS - Ma' mum Absi-Halabi 

ELECTROCHEMICAL METHODS IN BIOINORGANIC CHEMISTRY - 33 

Steven L. Suib 

ELECTRON TRANSPORT IN CHLOROPHYLL - John R. Long 35 

STRUCTURAL AND ELECTRONIC PROPERTIES OF BIS (CYCLOPENTADIENYL) 38 
TITANIUM AND VANADIUM COMPOUNDS - Benjamin Fieselmann 

EXPERIMENTALLY DETERMINED RATES OF THERMAL ELECTRON TRANSFER 40 
IN MIXED-VALENCE IRON COMPOUNDS - Ronald G. Wollmann 

THE ADSORBENT AND THE ADSORBATE: APPLICATION OF EPR TO 42 
CATALYSIS - Muin S. Haddad 

REACTIONS OF THE UNSATURATED CLUSTER H 2 0s 3 (C0) lO WITH 45 

LEWIS BASES AND OLEFINS - Jerome Keister 

THE REACTIONS OF UNSATURATED TRIOSMIUM CARBONYL CLUSTERS - 46 
Mamoru Tachikawa 



-1- 

INVESTIGATI ONS INTO THE ELECTRONIC STRUCTURE OF DI OXYGEN COORDINATED 

TO COBALT (II ) COMPLEXES 

David J. Kitko (Final Thesis) May 25, 1976 

Introduction 

The reversible binding of molecular oxygen to transition metal com- 
plexes has been an active area of research for many years. In 1947, 
Calvin and Bailes reported on their extensive investigations of dioxygen 
binding to a series of tetradentate Schiff-base complexes of Cp(ll) in 
the solid state. 1 Since that time, a variety of ligand environments 
surrounding a Co(ll) ion has been shown to lead to reversible binding of 
dioxygen. More recently, the research in this area has been motivated 
by a desire to elucidate the factors influencing the binding of 2 in 
biological systems, and to understand the mechanisms of the highly se- 
lective oxidations carried out with 2 in some of these systems. Our 
primary interest in this area has been the activation of 2 in the oxi- 
dations of organic substrates, particularly olefins. It was felt that 
by coordinating dioxygen to a transition metal center, it would be pos- 
sible to overcome some of its kinetic inertness, and perhaps obtain 
some degree of selectivity in the oxidation. As basic information, we 
thought it necessary to determine the net perturbation made on dioxygen 
upon coordination. The Co(ll) complexes were ideal for this study as 
they, in general, result in 2 adducts with one unpaired electron making 
them suitable for study via ESR. A great deal of research had previously 
been done on 2 adducts of Co(ll), and these studies led to their charac- 
terization as Co(lll)*0 2 . 2 For a number of reasons, it was felt that 
this characterization, although correct in the formal sense, did not pro- 
vide a true description of the perturbation made on the 2 fragment upon 
coordination. In an attempt to clarify this problem and to gain some 
additional information concerning the stability of the 2 adducts, the 
following studies were conducted. 

Results and Discussion 

Earlier studies had indicated that Co(SAIMeDPT) , a neutral penta- 
coordinate Schiff-base complex, reversibly bound 2 in solution at low 
temperatures. 3 However, attempts to use this complex in the catalytic 
oxidation of a variety of olefinic substrates failed. Systematic vari- 
ations in the ligand framework were then implemented. The N-CH 3 group 
was replaced by an ethereal oxygen donor, -0-, and thioether donor, -S-, 
and substituents were added to the aromatic rings. As a final pertur- 
bation, the salicylaldehyde portion of the Schiff-base ligand was re- 
placed by the 3-ketoamino structure through the condensation of the 
primary amines with acetylacetone and benzoylacetone. 

The reactivity of these complexes towards Lewis bases and towards 
2 was then investigated using ESR and UV-visible spectroscopy. Low 
spin adducts formed between all of these complexes and strong rr-back- 
bonding Lewis bases, and were characterized by ESR. The Co(X-SALDAPE) 
and Co(acac 2 MDPT) complexes also reversibly bound 2 . The 2 adducts 
were also characterized by ESR in frozen toluene/CH 2 Cl 2 solution. The 
very large cobalt hyperfine parameters found for the adducts of 
Co(X-SALDAPE) series coupled with a report of the anisotropic 1T hyper- 
fine in the ESR spectrum" of Co(bzacen)py 1T 2 4 led to a qualitative 
molecular orbital scheme to explain the bonding between Co and the 2 



-2- 

fragment. This model predicted that the unpaired electron will reside 
on the 2 fragment regardless of the extent of electron transfer from 
Co to 2 . A detailed analysis of the anisotropic cobalt hyperfine, using 
a spin polarization analysis similar to that employed elsewhere by 
Raynor 5 and Symons, 6 has led to an estimate of the extent of electron 
transfer from Co to 2 upon adduct "formation. The extent of electron 
transfer roughly parallels the strength of the donor atom set 
surrounding c oba It . Y 

A KINETIC STUDY OF A METAL ION COORDINATION SPHERE EXCHANGE REACTION 

In recent years, there has been a growing interest in polynuclear 
transition metal complexes. This research has been spurred by a desire 
to obtain fundamental information about two very important areas, 
metalloenzymes and heterogeneous catalysis. The general thrust of these 
research efforts has been to prepare simpler compounds which model in 
structure and/or reactivity of the active site of the in vivo metallo- 
enzyme or the active site of a heterogeneous catalyst. Our research in 
this area has been motivated by a desire to determine and attempt to 
understand the effect of one metal center on the chemistry (reactivity 
and physical properties) of the second metal center within the polynuclear 
compound. ! Bryan has recently prepared and characterized a series of 
binuclear complexes of the type M(salen)M' (hfac) 2 . 8 In the process of 
^characterizing these compounds, it was discovered that the binuclear 
complex Cuf salen)Co(hfac) 2 could be prepared by the direct reaction of 
Cufsalen) with Co(hfac) 2 « 2H 2 or by a ligand interchange process using 
Co(salen) and Cu(hfac) 2 , equation 1. 

Co(salen) + Cu(hfac) 2 -> Cu( salen)Co(hfac) 2 (l) 

The latter reaction was very rapid at 25°C in dilute CH 2 C1 2 solution. 
This result was intriguing and unexpected in this solvent medium where 
ligand dissociation and the generation coordinatively unsaturated ionic 
species is regarded as highly unlikely. Information about the mechan- 
ism of this unique metal ion-coordination sphere exchange process was 
considered necessary in attempting to understand the chemistry of this 
class of binuclear complexes. A detailed kinetic investigation of this 
reaction was therefore conducted using stopped-flow, UV-visible 
spectroscopy. 

Results and Discussion 



The kinetics of the reaction between Co(salen) and Cu(hfac) 2 in 
CH 2 C1 2 at 25°C were studied with Cu(hfac) 2 present in 3- to 8 0- fold 
excess. The reaction was monitored at 470nm, which corresponds to a 
flat region in the visible spectrum of Co(salen) and a minimum in the 
Cu(hfac7 2 spectrum. Even under conditions of ca. 8o-fold excess of 
Cu(hfac) 2 , the overall reaction does not exhibTE" pseudo-first-order 
kinetic behavior. The kinetic curves indicate the very rapid formation 
of an intermediate, which proceeds to product at a much slower rate. 
Rate studies at low Co(salen) and Cu(hfac) 2 concentrations indicate 
that the intermediate is generated by a second-order process, first- 
order in each reactant. The further reaction of the intermediate appears 
to be catalyzed by Cu(hfac) 2 . The rate of disappearance of the inter- 
mediate was subjected to a pseudo- first-order kinetic treatment. A 



-3- 

plot of k -. vs. Cu(hfac) 2 concentration yields a straight line with 

slope = 23 M" 1 sec" 1 and intercept = 0.53 sec" 1 . This information 
suggested the following mechanism for the reaction. 

Co(salen) + Cu(hfac) 2 ^ Co(salen)cu(hfac) 2 (2) 

Co(salen)Cu(hfac) 2 > Cu(salen)Co(hfac) 2 (3) 

Co£aler}Cu(hfac) 2 + Cu(hfac) 2 -> Cu^alet)Co(hfac) 2 

+ Cu(hfac) 2 (4) 

In order to verify the suitability of this mechanism, the 
PLATO IV computer system was used to calculate theoretical curves 
of aT _vs. time via numerical integration of the differential equations 
corresponding to a proposed mechanism. Plots of experimental data were 
superimposed upon the calculated curves in order to estimate the ac- 
curacy of the fit. 

It was possible to fit the kinetic curves for a wide range of 
concentrations of reactants with the mechanism represented by 
equations 2-4. Further kinetic and visible spectroscopic studies lend 
support to a mechanism in which the coordination sphere exchange pro- 
cess occurs without dissociation of a ligand, and indicates that initial 
adduct formation is required before the exchange will occur. 

References 

1. R. H. Bailes and M. Calvin, J. Amer. Chem. Soc, 69, 1886 (1947). 

2. B. M. Hoffman, D. L. Diemente and F. Basolo, ibid . , 92 , 6l (1970). 

3. B. S. Tovrog and R. S. Drago, ibid . , 96, 6765 (1974). 

4. D. Getz, E. Melamud, B. L. Silver, and Z. Dori, ibid . , 97, 
3846 (1975). ~~ 

5. B. A. Goodman and J. B. Raynor, Adv. Inorg. Chem. and Radiochem., 
135 (1970). 

6. T. F. Hunter and M. C. R. Symons, J. Chem. Soc. (a), 1770 (1967). 

7. B. S. Tovrog, D. J. Kitko and R. S. Drago, J. Amer. Chem. Soc, 
accepted. 

8. N. B. 0' Bryan, Ph.D. Thesis, University of Illinois. 



-4- 



TEMPERATURE DEPENDENT STUDIES OF STRUCTURAL AND ELECTRONIC 

EFFECTS IN SINGLE CRYSTALS 

Gerald Delker (Final Seminar) May 28, 1976 

The solid state properties of two transition metal systems which 
exhibit magnetic moments that change abruptly as a function of tempera- 
ture are reported. The first system is the octahedral complex tris(4- 
( (6-methyl)-2-pyridyl)-3-aza-3-butenyl)amino iron(ll) hexafluorophos- 
phate, which exhibits a sharp change in magnetic moment from high-spin 
to low-spin near 215°K in the solid state. In order to investigate 
this transition, the single crystal x-ray structure, I.R. spectra and 
the visible spectra were determined both above and below the transition. 
The Fe-N bond lengths determined for the 295°K crystal structure are 
consistent for an iron atom in the high-spin configuration, while at 
8o°K, a decrease in the Fe-N bond lengths of 0.17$ is observed which is 
consistent for a low-spin structure. The I.R. and visible spectra are 
observed to change as a function of temperature, with the I.R. spectra 
reflecting the changes in the Fe-N and N=C vibrations agreeing with a 
change in bond strengths. To investigate the effect that an anion might 
have on the magnetic susceptibility, the complex was prepared using 
tetraf luoroborate, t et raphe ny lb orate, bromide, and iodide anions. The 
magnetic susceptibility, I.R. and visible spectra were investigated for 
each of these complexes, and compared with the hexafluorophosphate com- 
plex. It is observed that the size of the anion has an effect on the 
spin transition, such that for larger anions, and for complexes which 
have solvent molecules trapped in the lattice, the complex remains high- 
spin, or has only a slight decrease in magnetic susceptibility, while 
for smaller anions, the complex is low- spin at room temperature. These 
results are discussed in terms of intermolecular Interactions via phonon 
coupling to the lattice. The second system is the one-dimensional semi- 
conductor tetrathiofulvalinium bis-cis- (l, 2-perfluoromethylethylene- 
1, 2-dithiolato) copper (i ) , which exhibits a small, but sharp break in the 
magnetic susceptibility near 250°K due to a first order phase transition, 
and undergoes a spin-Peierls transition to a singlet ground state at 
12.4°K. In order to investigate the high temperature transition, the 
single crystal x-ray structure was determined at 295° and 200°K. It 
is observed that the transition involves the freezing out of the ro- 
tation of the CF 3 groups accompanying a shift in the stacking of the 
molecules. The low temperature transition was investigated using ac 
calorimetry in order to measure the thermodynamic values of the trans- 
ition and to compare them with the values predicted by theory. 



-5- 

TH£ MECHANISM OF NI (I) -CATALYZED OLEFIN ISCMERIZATION 

i4icnael J. D'Aniello, Jr. (Final Tnesis) June 15, 1976 

Solutions of the tfi(I) complexes (PPh 3 ) 3 NiX, X=Cl,Br ,SnCl 3 , 
effectively catalyze the isomerization of terminal olefins to internal 
olefins witn an initial hign selectivity for the cis isomer that appears 
to be anion-dependent. Since it was not obvious that these complexes 
effected tne conversion by estaolisneci mechanisms a detailed investiga- 
tion was undertaken to determine the process by which tnese Ni(I) com- 
plexes induced isomerization. The proolera was approached in three ways 
and the results of each are outlined below. 

First, tne kinetics of the reaction of 1-butene with (PPh_) 3 NiSnCl-. 
were studied in botn oenzene and THF. Unfortunately, a rate law could 
not be determined due to poor reproducibility among kinetic runs. How- 
ever, several kinetic trends were apparent and must be accounted for by 
any proposed mecnanism. A nearly second order dependence was most often 
observed on [ (PPh-J ^IliSnCl^] . A zero order dependence was found for 
1-butene over tne concentration range of 0.01-0.5 M. There appeared 
to be an inverse dependence on added PPn.. but an order could not be 
determined. The rate of isomerization in benzene was greater than 
in THF whicn suggests a dependence on tne donor properties of the solvent. 

Another approacn to the problem was to investigate the isomeriza- 
tion of l-butene-3-d ,. Intermolecular scrambling of deuterium is 
generally observed with catalysts uased on metal hydrides while a 
specific 1,3-shift of deuterium occurs in systems which involve a 
TT-allyl metal hydride intermediate. The isomerization of l-butene-3-d 2 
was carried out with (PPh 3 ) -NiSnCl 3 in benzene and allowed to proceed 
to various levels of conversion. Tne unisomerized l-outene-3-d 2 and 
its isomerization products were isolated, separated, and examined by 
proton WMR and mass spectroscopy. The observed intermolecular scrambling 
of deuterium is consistent only with a nickel hydride based mechanism. 
At low conversion, net transfer of deuterium from tne 2-outenes to 
l-butene-3-d 2 occurred wnicn resulted in the formation of predominately 
mono-deuterated 2-butenes. After prolonged contact of l-butene-3-d 2 
witn (PPn«) ^NiSnCl^ tiiere was ooth an equal distribution of deuterium 
among isomers as well as a statistical distribution of deuterium within 
each isomer. 

As a result of tne laoeled olefin studies which indicated tnat 
a nickel nydride was responsible for isomerization, several experiments 
were designed to define tne nature of tne hydride intermediate and the 
process by whicn it formed from tne Ni(I) precursor. Treatment of 
(PPh 3 ) 4 Ni witn HC1 or HSnCl 3 at -78*C gave thermally sensitive hydrides, 
possibly HNi(PPn.J 3 X, wnich were very efficient isomerization catalysts 
and gave product ratios similar to those obtained with (PPn 3 ) 3 NiCl and 
(PPh 3 ) ^NiSnCl.,, respectively. These Ni(0)/acid systems were very 
sensitive to poisoning r>y 1,3-butaaiene wnicn reacted witn tne hydride 
to give a stable TT-crotyl nicKel complex. The Ni(I) systems were 
similarly affected. Addition of 0.01 mole of 1, 3-butadiene per mole 
of (PPn.J -jNiSnCl., stopped olefin isomerization. When the metal hydride 
formed from Ni (0]T/acid was treated witn 0.5 equivalent of 1 , 3-outadiene 
the following reaction occurred: 

2 HNi (PPn ) ^X + \fl * butenes + 2 (PPh ) NiX 



-o- 



wnich, if reversible, would provide a source of nickel hydride in 
the Ni(I) isomerization systems, i.e., 



-2P 



2 (PPh 3 ) 3 NiX 



•-► 2 (PPh 3 ) 2 NiX 



•-»■ HWi(PPh 3 ) 3 X + C-Ni(PPh 3 )X 



Support for this hypotnesis came from the traces of ir-crotyl nickel 
complex that were detected in Ni(I) isomerization reactions. 

Formation of the nickel hydride in tnis fasnion provides access 
to a standard catalytic cycle based on the metal hydride. The mechanism 
proposed on the basis of tne above data is outlined below. The source 
of cis selectivity can be rationalized by a consideration of the steric 
requirements of the olefin metal complex intermediates. 



2P 3 NiX 

-2P 

2P 2 NiX 



HNiP 3 X + tf-NiPX 




NiP 2 X 



8 / HNiP 2 X 



-7- 



STUDIES OF CYCLOPENTADIENYLBISLIGANDNICKEL(l) COMPLEXES 

Alex N. Williamson (Final Thesis) June 25, 1976 

A new class of cyclopentadienylbisligandnickel(l ) complexes has 
been prepared using a variety of ligands and several preparative methods. 
Ligands include amines, arsines, phosphites and phosphines. Preparative 
methods include reductions of cyclopentadienylnickel(ll) complexes, corn- 
prop ortionat ion of nickelocene and nickel (o) complexes and reactions in- 
volving oxidations of zerovalent nickel complexes. 

The paramagnetic products were detected and characterized by epr 
spectroscopy. Predictable hyperfine splitting patterns were observed at 
room temperature for species containing phosphorous and arsenic donors. 
Hyperfine structure was not observed at room temperature for species con- 
taining nitrogen donors. Frozen solutions of the amine complexes show 
well resolved hyperfine structure, as do the complexes with phosphines. 
Additionally, rhombic spectra were observed for frozen solutions of both 
the amine and phosphine containing species and reflect the low symmetry 
of these complexes. Computer simulated spectra were obtained for amine 
and phosphine complexes and are in close agreement with those observed 
experimentally. 

Complexes of suitable purity were isolated when 1, 2-bis(diphenyl- 
phosphino) ethane (diphos), 1,1' bipyridyl (bipy) and di-n-butylphenyl- 
phosphine were used as ligands. Magnetic susceptibilities were deter- 
mined for these complexes and are in the range of 1.69 to I.76 B.M. 
Attempts to isolate others of these species were generally unsuccessful 
and, in some cases, Ni(o) and Ni(ll) products were observed. Solutions 
were deep blue or violet in color, except those containing phosphites 
and bulky phosphines. The observation of brownish green solutions in 
these cases suggested that decomposition of the nickel (i) species 
was occurring. 

To obtain information about the stability of (C 5 H 5 )NiL2 systems, 
several cationic cyclopentadienylbisligandnickel(ll) complexes were 
studied by cyclic voltammetry and controlled potential electrolysis. Al- 
though strict electrochemical reversibility was not observed for most of 
these systems, peak current ratios (l p /lp) suggest that several of the 
Ni(l) species generated in this manner are stable even at slow potential 
scan rates. Current ratios observed for the Ni(l) complex with bulky 
triphenylphosphine show that this species is unstable at fast scan rates. 
Controlled potential electrolysis of several of these systems shows 
that the Ni(l) complexes once generated undergo extensive decomposition. 
One of the electroactive decomposition products was shown to be nickel- 
ocene by the process of peak matching. 

Chemical reactions of complexes with diphos and di-n-butylphenyl- 
phosphine were studied. Reactions with carbon monoxide resulted in dis- 
placement of the cyclopentadienyi ring and the formation of (CO) 2 NiL 2 
complexes. A transient (C 5 H 5 )NiL 2 L' species was detected (epr) in the 
reaction with P(0CH 3 ) 3 and suggests that addition of ligands and subse- 
quent displacement of the cyclopentadienyi ring occurs in a stepwise 
manner. 



ELECTRON EXCHANGE INTERACTIONS IN COPPER(ll) AND MANGANESE (II ) 

OUTER-SPHERE-ASSOCIATED DINERS ; THE DYNAMICS OF ELECTRON TRANSFER 

Edward J. Laskowski (Final Thesis) July 6, 1976 

It has been observed that the rate of certain outer- sphere redox 
reactions involving cationic transition metal complexes is affected by 
anionic species in solution. 1 * 2 This "anion assistance" effect gives 
rise to a term in the rate law for these reactions which is approxi- 
mately first order in concentration of added anion and dependent on the 
specific anion present. For those outer-sphere redox reactions observed, 
it was suggested that a precursor complex was formed in which the anion 
is complexed with the reducing agent and participates in the electron 
transfer process by providing a relatively favorable "path" between the 
two metal ions. 2 If the precursor species is relatively long lived, 
then the rate of electron transfer between the two metals will become 
important in determining the overall rate of the reaction. Assuming 
the above interpretation to be valid, a study of the magnetic exchange 
interactions in a series of outer-sphere-associated metal dimers could 
help in further understanding the "anion assistance" effect. It was 
with that purpose in mind that the present study was undertaken. 

A series of compounds having the formula [Cu 2 (tren) 2 X 2 1 (BPh^ ) 2 , 



](B 

) 3> 



where X is a halide or pseudohalide and tren is N(CH 2 CH 2 NH 2 ) 3 , has been 
prepared. The compounds have been structurally and magnetically charac- 
terized. For the compounds with_X = CN~ , 3 NCO , 4 NCS , 4 and Cl", 4 the 
structures contain discrete BPIl*" anions and dimeric cationic units, 
as shown by single-crystal X-ray diffraction studies. The two halves 
of the dimeric unit are connected by an outer- sphere type of bridge in 
which the X group is bonded to one of the copper (il) ions and also 
hydrogen bonded to one of the amine nitrogen atoms coordinated to the 
second copper (il) ion. Evidence for similar structures for the bromo- 
and hydroxo-bridged species is also given. It was necessary to establish 
that all compounds in the series contain copper ions in similar geometric 
environments so that complications due to differing ground states of the 
metal ion would not have to be considered. It has been found that by 
varying the X group, the magnitude of J, the electron exchange parameter 
in the spin Hamiltonian }4- ■--- 2JS T *S 2 , may change by as much as a factor 
of 100. Changes in outer-sphere redox rates of this order or larger 
have been observed for the "anion assistance" effect. The results of 
this study are discussed in terms of the relationship of J to the rate 
of electron transfer in outer-sphere redox reactions. 

Additional work in developing this possible solid-state analogue 
of the precursor in outer-sphere redox reactions has involved changing 
the metal from copper(ll) to manganese (il) . X-ray crystallographic 
studies show that [Mn 2 (tren) 2 (NC0) 2 1 (BPh 4 ) P and [Mn 2 (tren) 2 (NCS ) 2 ]- 
(BPh 4 ) 2 are isostructural with [Cu 2 (trenJ 2 (NCS) 2 ] (BPh 4 ) 2 . Detailed 
magnetic susceptibility and EPR work, including a single crystal EPR 
study, show these compounds to be dimeric, although antiferromagnetic 
exchange interactions of less than 0.20 cm 1 are found. The diffi- 
culties associated in obtaining precise values of J for these complexes 



make further generalizations about the dynamic nature of outer- sphere 
electron exchange interactions impossible. 

References 

1. J. K. Yandell, D. P. Fay, and N. Sutin, J. Amer. Chem. Soc, 95, 

1131 (1973). 

2. T. J. Przystas and N. Sutin, J. Amer. Chem. Soc, 95, 5545 (1973) 

3. D. M. Duggan and D. N. Hendrickson, Inorg. Chem., JL3, 1911 (197*0. 

4. E. J. Laskowski, D. M. Duggan, and D. N. Hendrickson, Inorg. Chem., 

14, 2449 (1975). 



-10- 

NOVEL STUDIES OF TRANSITION METAL DIMERS 

■ 

Robert M. Richman (Final Thesis) July 8, 1976 

- 

Part 1. Physical Studies of Rhodium Carboxylate Dimers. 

The chemistry of compounds containing two or more transition metal 
ions has recently gained widespread attention. In an attempt to under- 
stand the acid-base and redox properties that make these systems at- 
tractive as biological and industrial catalysts, a series of physical 
measurements has been performed on the rhodium butyrate dimer and its 
one-to-one and two-to-one adducts with several Lewis bases. Isosbestic 
points in the electronic spectra indicate that in most cases the first 
mole of base binds completely before the second mole binds, and that 
both steps have high equilibrium constants. This facilitates the de- 
termination of the enthalpies of adduct formation of the separate steps. 
Using seven bases with known E„ and C R parameters, it is not possible 
to obtain adequate E. and C„ parameters for rhodium butyrate. This is 
explained in terms ox tt effects. These results motivated an electro- 
chemical investigation of the adducts, since oxidation corresponds to 
the removal of an electron of tt symmetry. Indeed, the measured E x / 2 
values are very sensitive to the nature of the coordinated base. The 
spectroscopic and electrochemical results can be combined with the 
calorimetric results via thermodynamic cycles to estimate enthalpies of 
adduct formation of the rhodium butyrate ion and excited state with the 
bases studied. In the cyclic voltammetry experiment, it was noted that 
the peak potential separation differed markedly for different adducts. 
After correcting these data for iR drops within the cell, it has been 
possible to extract rate constants for electron transfer at the platinum 
electrode and draw some tentative conclusions about the mechanisms of 
electron transfer in the various adducts. 

In a companion study, the similar but stronger acid rhodium tri- 
f luoroacetate has been allowed to interact with the nitroxide radical 
2, 2, 6,6-tetramethylpiperidinc-N-oxyl. The X-band and Q-band esr spectra 
show the presence of a one-to-one complex with rhodium hyperfine of 
1.7 J_- 0.2 gauss and a g shift from 2.0052 to 2.0150, the largest ever 
reported for a nitroxide radical. These results are interpreted semi- 
quantitatively in terms of rhodium's large spin-orbit coupling constant 
and a molecular orbital model that requires delocalization of the un- 
paired electron into the tt* combination of rhodium d, orbitals. 

Part 2. The Rate of Thermal Intervalence Transfer in u-pyrazine- 
decaa.mmineuiruthenium(5 ) . 

The title compound has boon one of the most thoroughly studied 
mixed valence compounds, yet there is no agreement as to even the ap- 
proximate time scale at which the odd electron changes from a "localized" 
to a "delocalized" description. It is proposed that this rate of 
intervalence transfer in this and similar compounds can be determined 
by measuring the frequency dependence of the dipole moment in a frozen 
solution of the compound. Such measurements of dielectric permittivities 
have been done for many years by frequency domain techniques. In the 



-11- 



recently developed technique of time domain ref lectometry, the di- 
electric response of the sample to a voltage pulse is Fourier trans- 
formed, yielding information over several orders of magnitude 
in frequency. 

Preliminary results indicate a rate of 2.5 + 1.0 x 10 9 sec" 1 at 
250°K. The theoretical basis for estimating these rates from the Hush 
model is critically examined and greatly modified by a demonstration 
of the importance of quantum mechanical tunneling and anharmonicity 
in the potential energy. 



-12- 

ELECTRON EXCHANGE INTERACTIONS TN DTMERIC LINEAR BPTDGED 

COMPLEXES OF CU(IT) AND IN DTBRTDGED COMPOUNDS OF TI(TTI) 

Rudolph G. Jungst (Final Thesis) August 2h 9 1976 

The magnetic properties of a series of macrocyclic ligand copper 
complexes containing a single bridging group have been studied to 
determine the strength of the electron exchange interaction. A cya- 
nide dimer of this type, {[Cu(MeA(l I f)- l +,ll-dieneNi + )] 2 CN}(ClO^) 3 , was 
first synthesized by Curtis. 1 The structure of the macrocyclic lig- 
and is shown below. More recently, Bauer, Robinson, and Margerum 




-VY 



tet-b 



J Me 6 (l^-)- I +.ll-dieneNi + f J 

prepared and determined the X-ray crystal structure of {[Cu (tet-b)] 2- 
Cl} (010^)3 which has a single bridging chloride. 2 This series has 
now been extended by the synthesis of -{[Cu(tet-b)] 2 CNJ(C10k ) 3 and 




simulated by an exchange term of the form "tf = 2JS-pS2 with J gauging 
the magnitude of the interaction. Results from computer fitting the 
data are J = -3 cm"l for the cyanide dimer with the diimine ligand 
and J = -lV+ cm~l f -23 cm"l, and -1*+ cm" 1 for the chloride, cyanide, 
and azide dimers with tet-b. An X-ray crystallographic determination 
confirms that the cyanide in the diimine ligand complex is bridging 
and infrared spectra are consistent with end to end bridging in the 
ON" and N3" tet-b complexes. Electronic spectra and epr studies 
indicate that the copper site has similar properties for the chloride 
and cyanide tet-b dimers, while a different distortion than has been 
previously investigated exists in the cyanide dimer with the diimine 
ligand. The difference in J for the two cyanide dimers is attributed 
to the change in ligand geometry about the copper atom while the tet-b 
chloride and cyanide compounds provide a good comparison of the effec- 
tiveness of exchange coupling through these two bridges without simul- 
taneous-effects from distortion of the nonbridging ligand geometry. 




for [(C 5 H 5 )2Ti]2MnCl lf (THF) 2 




-13- 

also 

icrease 
>graphic 

data indicate that the increase in J parallels a decrease in metal- 
metal separation, implying direct overlap of metal orbitals as a 
possible exchange pathway. evidence is seen in the variable temper- 
ature magnetic susceptibility data on [(C5H5) 2 TiCl] 2 for impurities 
which are introduced during sublimation. 

» 
References 

1. Y. M. Curtis and M. F. Curtis, Aust. J. Chem., 12, 609 (1966). 

2. R. A. Bauer, W. R. Robinson and D. W. Margerum, J. Chen. Soc., 
Chem. Commun., 289 (1973). 

3. D. G. Sekutowski and G. D. Stucky, Tnorg. Chem., }}± $ 2192 (1975). 



-14- 



THE IMPORTANCE OF SYMMETRY IN THE UNDERSTANDING OF PHASE CHANGES IN 
CERTAIN THREE-DIMENSIONAL ANTIFERR (MAGNETS 

Jeffry A. Kelber November 9, 1976 

A phase change can be defined as a change in the symmetry or 
structure of a crystal at a given temperature or pressure. Phase 
changes which occur in such compounds as TTF:TCNQ, K 2 Pt (CN) 4 Br . 3 3H 2 0, 
and various magnetic systems occur as a function of temperature only. 
Therefore, tr'ansitions which occur at a given critical temperature 
are of current interest to chemists. 

Phase changes can generally be classified as first or second 
order, depending on whether or not the transition between phases is 
discontinuous or continuous. 1 The mean field theory developed by 
Landau 2 fails to correctly predict the first-order transition in 
several three-dimensional antiferromagnets. 3 Mukamel, Krinsky, and 
Bak, 3 ~ 5 using renormalization group methods, have developed a new 
criterion for predicting the order of the phase transition in certain 
compounds. This theory, based upon symmetry considerations rather 
than upon detailed physical interactions within the lattice, corrects 
the discrepancies of the Landau theory and predicts that the trans- 
itions of several other compounds are first order. 3 Bak also predicts 
that impurities in these compounds will lead to a gradual rounding or 
smearing of the discontinuity of the first order transition. 6 

Recently experimental work has been done on two of the compounds 
predicted to undergo first order transitions: TbSb 7 and ErSb. 8 " 10 
While the results are not conclusive, the weight of the evidence 
currently supports the new theory. 

REFERENCES 

Rao and K. Rao, Prog, in Solid State Chem. , k, 131 (1967). 
Landau and E. Lifshitz, Statistical Physics, Pergammon, N.Y. 

(1968). 

Mukamel, S. Krinsky and P. Bak, Phys. Rev. Letters, 36, 

1 (1976). 

Mukamel and S. Krinsky, Phys. Rev. B, 13, 5065 (1976). 

Mukamel and S. Krinsky, Phys. Rev. B, T3, 5078 (1976). 

Bak, to be published. 

Carneiro, N. Andersen, J. Kjems and 0. Vogt, to be published. 

M. Shapiro and P. Bak, J. Phys. Chem. Solids, 36, 579 (1975). 

Hulliger and B. Natter, Solid State Comm. , 13, 221 (1973). 

Cox, et al, to be published. 



1. 


c. 


2. 


L. 




( 


3. 


D. 


4. 


D. 


5. 


D. 


6. 


P. 


7. 


K. 


8. 


S. 


9. 


F. 


10. 


D. 



-15- 

HETEROGENEOUS CATALYSIS EY ZEOLITES 

David J. Blumer November 11, 1976 

Introduction 

In i960, Weisz and Frilette 1 disclosed the first application 
of zeolites to catalysis by selectively cracking n-paraffins with 
NaX and CaX zeolite. This discovery spurred an intensive research 
and development effort which has led to the commercial use of metal- 
substituted zeolites to crack over four million barrels of petroleum 
per day. 2 This research has shown that the catalytic and physical 
properties of zeolites can be extensively modified, allowing chemists 
to "engineer" a highly selective catalyst for an extremely broad 
range of reactions. 3 

Zeolite Composition 

Zeolites are crystalline aluminosilicate compounds which form 
porous, three-dimensional frameworks of T0 4 (T=Si, Al) tetrahedra 
which are linked through the oxygen atoms to form polyhedra which 
interconnect to form channels and cages. Charge balance is provided 
by cations located in several different environments within the frame- 
work. In general, the cations can be exchanged by other cations as 
long as steric and electronic requirements are fulfilled. Some 34 
naturally-occurring zeolites have been identified while over 90 syn- 
thetic zeolites have been characterized. The most common zeolites 
which are used as catalysts are the synthetic faujasites NaX and NaY 
which have a range of stoichiometrics Na 7 y_ 96 (A10 2 )77-96 (Si0 2 ) x 20-100- 
• nH 2 and Na 4 9_ T6 (A10 2 ) 49 _7 6 (Si0 2 ) 143-1 21 • nH 2 , respectively, having 
pore openings of 8-10 A. 3 ' 4 The sodium ions can be exchanged for a 
great variety of cations, including transition metals, lanthanides, 
and organic cations. NaX and NaY are synthesized from aqueous gels 
of NaA10 2 , sodium silicate, NaOH, and colloidal Si0 2 at about 100 °C. 
The products obtained are dependent on the concentration of water, 
the Si/Al ratio, the temperature, pressure, and reaction time. 5 

Catalytic Facility 

A great deal of research has been directed toward discovering 
the variables which control the catalytic properties of zeolites. 3 
Some of these are the zeolite pore size and shape, the cations present 
and their oxidation state, the state of hydration, the Si/Al ratio, 
impurities in the framework (e.g., Fe 3 , Ga 3 ", Ge 4 ), and the tempera- 
ture treatment of the zeolite. An example of this is demonstrated 
by NaX, which has no activity in the hydrogenation of olefins, and by 
NiX, which facilly hydrogenates olefins, including phenol to give 
cyclohexanol. 6 

A wide variety of organic reactions can be catalyzed with high 
selectivity by employing the properly modified zeolite. Because of 
their size and shape selectivity, high activity, broad range of re- 
actions, and ease of modification, zeolites have been compared to 
enzymes. However, zeolites have the advantage of being stable over a 
broad range of reaction conditions. 3 



-16- 

Physical Measurements 

The nature of the active sites in zeolites has been investi- 
gated by many techniques including ir, nmr, and epr spectroscopy. 7 
Unfortunately, much of this work is of little value because of the 
lack of adequate purification and characterization of materials. 8 * 9 
Most workers have relied on x-ray powder diffraction and elemental 
analysis for characterization, but these techniques may not be sensi- 
tive enough to detect all the important variables such as trace metal 
impurities and the distribution of cations in the various sites in 
the zeolite. 

A technique for growing large single crystals of zeolites has 
recently been developed by Seff and coworkers 10 and they have reported 
the first high-resolution single-crystal x-ray results of adsorbed 
molecules in substituted zeolite A. When the zeolite is dehydrated, 
the metal ions become highly coordinatively unsaturated and can bind 
ligands such as acetylene, CO, and NO to give structures which appear 
similar to complexes well known in organometallic chemistry: C 2 H 2 
binds side on with M-C=2.6 A; CO binds linearly with M-C=2.3 A; NO 
gives a bent complex with M-N-0 bond angle of l4o° and M-N=2.4 A and 
N-0=1.5 A. No results have been reported on the catalytically more 
interesting zeolites X and Y. 

In a recent x-ray photoelectron study of nickel substituted NaY 
zeolites, Minachev, _et. al., 11 observed that upon reaction with CO 
at 400 °Q, Ni(0) and an intermediate form of Ni which they referred 
to as Ni" 1 * were obtained. They attributed the stabilization of Ni 
to the formation of Ni complexes with adsorbed CO. 

Catalysis Mechanism 

A number of workers have postulated that small metal clusters 
form in the supercages of transition metal substituted zeolites upon 
reduction by hydrogen at high temperatures. 12 ' 13 ' 14 Beyer, Jacobs, 
and Uytterhoeven 14 recently studied the kinetics of hydrogen reduction 
of AgY zeolite using ir spectroscopy and x-ray spectrometry. They 
concluded that the silver ions migrate out of the smaller cages into 
the supercages to form Ag 3 clusters, while at very high temperatures, 
Ag crystallites formed on the outside of the zeolite framework and 
were observable by electron microscopy. The evidence for the for- 
mation of metal clusters is not yet conclusive, but some interesting 
analogies from homogeneous catalysis by metal cluster compounds may 
exist if that evidence is forthcoming. 

Conclusion 

In conclusion, zeolites are useful and interesting compounds 
which are very active heterogeneous catalysts. Since these are 
crystalline materials, chemists may be able to gain insight into the 
mechanisms of heterogeneous catalysis through the study of the intimate 
details of the structures, reactions, and physical properties of 
these materials. 



-17- 
References 

1. P. B. Weisz and V. J. Frilette, J. Phys. Chem. , 64, 382 (1964). 

2. D. W. Breck, "Zeolite Molecular Sieves," Wiley, N.Y., 1974. 

3. P. B. Venuto, Adv. Catal. , 19, 257 (1968). 

4. J. V. Smith, Adv. Chem. Ser. , 101, 171 (1971). 

5. R. M. Milton, U. S. Pat. 2,882,244 (1959); D. W. Breck, 
U. S. Pat. 3,130,807 (1964). 

6. British Pat. 1,257,607 (1972). 

7. Kh. M. Minachev and Ya. I. Isakov, Adv. Chem. Ser., 121 , 
451 (1973). 

8. J. Turkevich and Y. Ono, Adv. Chem. Ser., 102, 315 (1971). 

9. D. Michel and V. Rossiger, Surface Sci., 5^(2) , 463 (1976). 

10. K. Seff, Ace. Chem. Res., 9, 121 (1976); P. F. Riley, 

K. B. Kunz, and K. Seff, J. Am. Chem. Soc. , 97, 537 (1975). 

11. Kh. M. Minachev, G. V. Autoshin, Yu. A. Yusifov, and 

E. S. Shpiro, Rxn. Kin. and Catal. Lett., 4, 137 (1976). 

12. R. A. Schoonheydt, L. J. Vandamme, P. A. Jacobs, and 
J. B. Uytterhoeven, J. Catal., 43, 292 (1976). 

13. W. Morke, R. Vogt, and H. Bremer, Z. Anorg. Allg. Chem., 422 , 
273 (1976). 

14. H. Beyer, P. A. Jacobs, and J. B. Uytterhoeven, J. Chem. Soc, 
Faraday I, 72, 674 (1976). 



-13- 

pa 

MOSSBAUER spectroscopy applied to supported metal catalysis 

Lynn C Francesconi November 16, 1976 

INTRODUCTION 

Reactions catalyzed by transition metals and metal compounds are 
industrially important. The details of the catalyzed reactions are 
generally not understood completely. Recently, several physical tech- 
niques have been used to investigate catalytic reactions which occur 
on surfaces. The kinetics and mechanisms of the reactions and the 
nature of the surface before, during, and after the reaction are 
being studied. 

Several investigations of catalytic reactions have concerned metals 
supported on inert carriers, such as aluminum oxide, silica, magnesium 
oxide, and graphite to mention a few. Supported metal catalysts have 
the advantages of possessing high surface area, maintaining small parti- 
cles on the support, and requiring a small quantity of metal to prepare 
the catalyst. 

Since 1963, Mossbauer spectroscopy has experienced increasing use 
in the study of supported catalysts. The Mossbauer effect, a recoil 
free emission and absorption of gamma rays, was discovered in 1957 by 
R. L. Mossbauer during his graduate work. This effect is exhibited by 
44 nuclei, of which 5Y Fe and 119 Sn have been most widely studied. The 
parameters obtained in the Mossbauer experiment can be useful to charac- 
terize the electronic state and the symmetry about the metal in the 
supported catalyst, as well as the oxidation-reduction behavior and 
particle size distribution, which cannot be determined easily by con- 
ventional techniques. 1 

IRON OXIDE 

Mossbauer parameters have proved useful in characterizing supported 
iron oxide catalysts. The strength of interaction of iron with various 
supports has been correlated to the ease of reduction of the ferric 
species on the supports as monitored by Mossbauer spectroscopy. 2 This 
type of information is important in understanding the catalytic proper- 
ties of iron oxide on various supports and may aid in catalytic design. 
Mossbauer spectroscopy has been used to study chemisorption onto sup- 
ported iron oxide and the penetration of particles into iron oxide to 
give insight into intermediates found in catalytic reactions such as 
dehydrogenation or isomerization. 3 

BIMETALLIC SYSTEMS 

Bimetallic catalysts have evoked interest in catalysis studies 
because the reactivity and selectivity of the bimetallic species differs 
from the individual components. In the majority of studies, the bi- 
metallic catalysts have been in the form of powders, wires, and films 
Recently, supported bimetallic catalysts have been investigated. To 



-19- 

date, there are two Mossbauer studies considering supported iron 
alloyed with Pd or Pt. 4 ' 5 Mossbauer spectroscopy has enabled the 
workers to characterize the iron species on the support as to the 
chemical state, symmetry, and concentration of surface iron. Re- 
versible oxidation-reduction behavior is observed and this supports 
the fact that the iron is associated with either the Pd or Pt as 
clusters on the support. 

ZEOLITES 

Another example of supported metal catalysts which has been widely 
used in industry is that of transition metals exchanged into zeolite 
lattices. It is found that the catalytic activity and selectivity is 
specific to the particular lattice with its characteristic geometry 
and pore size. Mossbauer spectroscopy has served to locate the nucleus 
in the various channels, cages, and pockets of the zeolite by mcnitcring 
the adsorption behavior, oxidation-reduction behavior and by comparison 
of parameters with compounds of known coordination number and chemical 
state. 6 Workers have used information derived from Mossbauer experi- 
ments to rationalize the catalytic activity and selectivity of various 
iron zeolites. 

REFERENCES 

1. H. M. Gager, Catal. Rev.-Sci. Eng . , 11, 1 (1975). 

2. H. Hobert, Proceedings of the Conference on Applications of the 

Mossbauer Effect (Tihany 19^9) / pp. 525. 

3. Y. V. Maksimov, et al. , Dokl. Aka. Nauk. SSSR , 206, 1120 (1972). 
b. R. L. Garten, J. Catal ., 43, l8 (1976). 

5. C. H. Bartholomew, J. Catal ., 29, 273 (1973). 

6. M. Boudart, Ind. Eng. Chern. Funriam. , 12, 299 (1973). 



-20- 

HIGH POTENTIAL IRON SULFUR PROTEIN AND 
THE THREE STATE HYPOTHESIS 

Carol Iris Ashby November 22, 1976 

Introduction 

The high potential iron sulfur protein ( HP) of the purple sulfur photo- 
synthetic bacterium, Chromatium vino sum , is an electron transfer protein 1 
containing four iron and* four inorganic sulfur atoms. The cluster arrange- 
ment of the iron and sulfur atoms 2 is identical, within experimental error, 
to that found in the two iron-sulfur clusters of the eight-iron-eight-sulfur 
ferredoxin (ED) of Peptococcus aerogenes . However, the properties of the two 
protein types differ considerably. The redox potentials of HP and both 8-Fe-8-S 
and 4-Fe-^S ferredoxins differ by over 700 mV. 3 It is reduced FD which 
exhibits an epr spectrum, whereas it is the oxidized form of HP which exhibits 
an epr spectrum. 3 These observations have led to the proposal that the cluster 
exists in three oxidation states. 23 



c" 


C 


C + 


FD(red) 


FD(ox) 


[FD(s-ox)] 


[HP(s-red)] 


HP (red) 


HP(ox) 


Physical Studies 







Several physical techniques have been employed to examine the properties 
of HP and FD of Clostridium pasteurianum . Information about the magnetic prop- 
erties has been obtained for both proteins by proton magnetic resonance studies 
in the temperature range 5 to 40° c 1 *' 5 ' 6 ' 7 and, for HP, by direct susceptibility 
measurements at low temperatures. 8 ' 9 The pmr contact shift studies indicate 
that the iron sites in FD(red) follow the Curie law, that both HP (red) and 
FD(ox) behave in a non-Curie fashion, and that two types of iron sites are 
present in HP(ox). Magnetic susceptibility data are consistent with the pro- 
posed ant i ferromagnetic nature of the HP cluster at low temperatures. Electron 
paramagnetic resonance studies have revealed only a single axial-type signal 
for FD(red) lls 12 , whereas K-band studies of HP(ox) have revealed the presence 
of two equally populated sites of rhombic and axial symmetry. 9 The signal of 
HP(s-red) 13 is very similar to that of FD(red). Mossbauer studies lu 17 lend 
further support to the three-state hypothesis. The spectra of FD(ox) 16 and 
HP(red) 17 show similar isomer shifts and Zeeman behavior. The Mossbauer spectra 
of HP (ox), 15 > 17 i n agreement with the epr spectra, show the presence of two 
different types of iron sites. The FD(red) 16 and HP(s-red) 1 " spectra have 
similar isomer shifts but otherwise differ. 

Structure of the Proteins 

Structural determinations for HP and FD have revealed some fundamental 
differences in the proteins which may account for the extremely different 
characteristic properties. 10-22 The relative positions in the proteins of the 
cysteines 18 ' 19 which bind the Fe u S u clusters differ considerably, and this is 
reflected in the position of the cluster relative to the intramolecular inter- 
face. 18 A comparison of the oxidized and reduced forms of HP has revealed 



-21- 

changes in the cluster structure 18 which provides an explanation of the ob- 
served epr, pmr, and Mossbauer spectra. The possible role of tyrosine in 
the electron transfer process has been examined. 18 

Conclusion 

The results of the studies conducted to date appear to support the three 
state hypothesis first proposed by Carter 23,2U in 1972. The hypothesis has 
provided a viable explanation of the extremely different properties of the 
two protein types 
i 

References 

1. P. L. Dutton and J. S. Leigh, Biochim. Biophys. Acta , 314, 178 (1973) 

2. T. Herskovitz et al., Proc. Nat. Acad. Sci. , USA , 69, 2437 (1972). 

3. W. Lovenberg, ed. Iron Sulfur Proteins Acad . Press (1973). 

4. W. D. Phillips. M. Poe, C. C. McDonald, R. G. Bartsch, Proc. Nat. Acad . 
Sci., USA, 67, 682 (1970). 

5. R. H. Holm et al., J. Amer . Chem . Soc , 96, 2109 (1974). 

6. M. Poe, W. D. Phillips, C. C. McDonald, W. Lovenberg, Proc. Natl. Acad . 
Sci., USA , 65, 797 (1970). 

7. R. H. Holm et al, J. Amer. Chem. Soc , 96, 2109 (1974). 

8. H. B. Gray et al., J. Amer. Chem. Soc , 96, 6534 (1974). 

9. B. C. Antenaitis and T. H. Moss, Biochim. Biophys. Acta , 405, 262 (1975). 

10. R. B. Frankel et al., Biochem. Biophys. Res. Comm . , 58, 974 (1974). 

11. G. Palmer, R. Sands, L. E. Mortensen, Biochem. Biophys. Res. Comm , 23 , 357 (1966). 

12. W. H. Orme-Johnson, H. Beinert, Biochem. Biophys. Res. Comm ., 36 , 337 (1969). 

13. R. Cammack, Biochem. Biophys. Res. Comm ., 54, 548 (1973). 

14. D. P. E. Dickson, R. Cammack, Biochem. J . 143, 763 (1974). 

15. C. W. Evans, P. 0. Hall, C. E. Johnson, Biochem. J . 119 , 289 (1970). 

16. C. L. Thompson et al, Biochem. J . 139 , 97 (1974). 

17. D. P. E. Dickson et al, Biochem, J . 139 , 105 (1974). 

18. K. Dus, S. Tedro, R. G. Bartsch, J. Biol. Chem . 248 , 7318 (1973). 

19. J. N. Tsunoda, K. T. Yasunobu, H. R. Whiteley, J. Biolog. Chem ., 243, 
6262 (1968). 



-22- 

20. C. W. Carter et al. J. Biol. Chem . , 249, 6339 (1974). 

21. C. W. Carter et al. J. Biol. Chem ., 2*49, 4212 (197*0. 

22. E. T. Adman, L. C. Sieker, L. H. Jensen, J. Biol. Chem. , 2*48 , 3987 (1973) 

23. C. W. Carter et al., Proc. Nat. Acad. Sci., USA , 69, 3526 (1972). 

24. A. J. Ihomson, Biochem, Soc. Trans., 3, 468 (1975). 



-23- 



STRUCTURE OF CYCLOPHOSPHAZENES IN RELATION 
TO THEIR METAL COMPLEXES 
Gerald V. Rubenacker November 29, 1976 

INTRODUCTION 

Cyclophosphazenes are of the general formula (NPR 2 ) n where 
-R can be a wide variety of groups including halogens, amines, 
alkoxides, alkyls, and aryls. 1 Compounds have been characterized 
having ring sizes from 6 through 18 atoms. It is the reaction of 
ammonium chloride and phosphorus pentachloride which provides 
the most convenient route for synthesis of chlorophosphazenes , 
(NPC1 2 ) 3 „ s 5 , g They are themselves used as precursors in 
substitution reactions to replace the chlorines by other types of 
groups. 3 Phosphazenes have been used as inorganic polymers, 4 
as clathrates 5 in the separation of organic compounds, as fertil- 
izers, 6 and as flame retardent materials for textiles. 

BONDING IN CYCLOPHOSPHAZENES 

The phosphazene ring is known to be significantly more stable 
than can be accounted for by N-P single bonds. At the same time 
one must account for equal P-N bond lengths and a low barrier to 
rotation around the P-N bond. Four models each with its own 
support have been used to explain these facts. The simplest model 
consists of ionic interactions whereby the nitrogen atoms are 
assigned a partial negative charge and the phosphorus atoms a 
partial positive charge. Some recent ESR work supports this model 7 . 
In the second model Paddock postulates a "benzene-like" derealization 
where 7T-bonding between nitrogen p z and phosphorus d xz orbitals is 
suggested by increasing ionization potentials 8 and P-N bond ener- 
gies as ring size is increased. The third is Dewar's three center 
island model for p^-d-fr bonding 9 which is supported by uv spectral 
data and Faraday effect measurements. 10 The fourth model accounts 
for stabilization through P-P transannular bonding, and this possi- 
bly is suggested by CNDO/2 molecular orbital calculations. 11 

COMPLEXES OF PHOSPHAZENES 

Metal complexes of phosphazenes were found to form most easily 
with aminophosphazenes l 2 where coordination is at a ring nitrogen 
atom. This is consistent with basisity studies of the ring 
nitrogen atoms. 1 Protonation of aminotricyclophosphazenes has a 
pronounced effect on the ring bonding. An alternation of bond 
lengths is induced indicating the breaking of the ring conjuga- 
tion. 13 ' 114 These effects are also observed in the tetramers x 5 ' x 6 
as well as a pronounced closing of the N-P-N ring bond angles due 
perhaps to the additional electron withdrawal from the ring. 



_24- 
The molecule N 6 P 6 (NMe 2 )i 2 has been used as a macrocyclic 



1 7 



ligand to form a number of complexes. Structural studies have 
been done on a Cu(II) 18 and a Co(II) 19 complex with a chloride 
ion occupying a fifth and equitorial coordination site. The 
macrocycle folds to block the possibility of a sixth ligand 
coordinating. These compounds provide the opportunity to ob- 
serve both ring conformation and axial and equitorial coordination. 
i 
With N U P 4 (NMe 2 ) 8 W(C0) u 20 we have an example of coordination 
both to the ring and to the exocyclic amine as opposed to coor- 
dination by two ring nitrogen atoms as in N U P U (NHMe) 8 PtCl 2 . 2 x 
The tungsten complex allows an examination of exocyclic coor- 
dination and gives an example in which the phosphazene is not 
symmetrically bound. 

REFERENCES 

1. H. R. Allcock, Chem. Rev ., 72, 315 (1972). 

2. J. Emsley and P. B. Udy , J. Chem. Soc, A, 768 (1971). 

3. R. A. Shaw, Z. Naturforsch. B , 31, 64l (1976). 

4. H. R. Allcock, Science, 193 , 1214 (1976). 

5- K. R. Allcock, R. W. Allen, E. C. Bissell, I. A. Smeltz and 
M. Teeter, J. Am. Chem. Soc , 98, 5120 (1976). 

6. W. Wanek, Pure Appl. Chem . , 44, 459 (1975). 

7. S. P. Mishra and M. C. R. Symons , J. C. S. , Dalton , 1622 (1976). 

8. G. R. Branton, C. E. Brion, D. C. Frost, K. A. R. Mitchell 
and N. L. Paddock, J. Chem. Soc. , A , 151 (1970). 

9- J. P. Faucher, J. F. Labarre and R. A. Shaw, Z. Naturforsch, B, 
31, 677 (1976). 

10. J. P. Faucher, 0. Glemser, J. F. Labarre, and R. A. Shaw, 
C. R. Acad. Sci . , 279C , 441 (1974). 

11. D. R. Armstrong, G. H. Longmuir and P. G. Perkins, Chem. Comm . , 
464 (1972) . 

12. T. Moeller and S. G. Kokalis, J. Inorg. Nucl. Chem. , 25, 875 

(1963). 

13. A. L. MacDonald and J. Trotter, Can. J. Chem. , 52_, 734 (1974). 

14. H. R. Allcock, E. C. Bissell and E. T. Shawl, Inorg. Chem. , 
12, 2963 (1973). 



-25- 

15. J. Trotter and S. H. Whitlow, J. Chem. Soc . , A , ^55 (1970). 

16. J. Trotter and S. H. Whitlow, J. Chem. Soc., A , 460 (1970). 

17- H. P. Calhoun, N. L. Paddock, and J. N. Wingfield, Can. J. Chem. , 
53, 1765 (1975). 

18. N. C. Marsh and J. Trotter, J. Chem. Soc, A , 1^82 (1971). 

19. W. Harrison and J. Trotter, J. C. S., Dalton , 6l (1973). 

20. H. P. Calhoun, N. L. Paddock and J. Trotter, J. C. S. , Dalton , 
2708 (1973). 

21. H. R. Allcock, R. W. Allen and J. P. O'Brien, Chem. Comm. , 
717 (1976). 



-26- 

cj^-DIAMMINEDICHIX>ROPMTINUM(ll) AND ITS ANALOGS: 
THE SEARCH FOR A CURE FOR CANCER 

Peter A. Bellus December 2, 1976 

Introduction 

While the discovery that cis -Pt (NH 3 ) gClg had activity in bio- 
logical systems was accidental, the discovery of its anti-tumor 
properties w^ls the result of systematic study. 1 A major aim of 
present research efforts is to develop better anti-cancer drugs analo- 
gous to this compound. If this is to be accomplished in other than a 
random fashion, inquiries must be made into the mode of action of 
these compounds on cancer tissue. 

Biological Effects 

Early studies showed that c_is-Pt(NH 3 ) 2 C1 2 does not concentrate 
in tumors, 2 but it was recognized that the complex had an effect on 
cell division. Recent studies on tumor cells show that cell division 
is blocked in metaphase, but protein and RNA synthesis remain active. 3 
Aggarwal 4 notes a clear band around the nucleus, indicating the ab- 
sence of microfilaments (probably due to their depolymerization) . Al- 
though disruption of cell division has been proposed as the mechanism 
for the anti-tumor action of the complex, Rosenberg 5 has proposed that 
cis-Pt (NH 3 ) 2 C1 2 and its analogs activate an immune response against 
cancer cells. He cites, among other things, the correlation between 
the ability of a complex to induce lysis in lysogenic bacteria and 
its ability to inhibit tumor growth. It should be noted that cis - 
Pt(NH 3 ) 2 Cl 2 has been shown to repress the immune response in certain 
instances. 6 It is not yet clear which of the two mechanisms is more 
important in curing cancer. 



The Chemistry of cis - Pt (NH 3 ) 2 C1 2 



The complex cis -Pt (NH 3 ) 2 C1 2 is fairly reactive, undergoing dis- 
placement of the chloride ligands by a variety of nucleophiles ( e. g. , 
^actuation = 2-5 x 10 5 sec" 1 ). 7 However, the dichloro form is most 
liReiy -one active form present in the cell. 8 cis-Pt (NH 3 ) 2 C1 2 is known 
to react with amino acids and proteins, but these reactions do not 
seem to be of much importance in the anti-tumor activity of the com- 
plex. 9 ' 10 Interactions with DNA have been studied extensively. 
Interstrand 11 and intrastrand 12 * 13 have been proposed to explain the 
activity of the complex, although certain evidence suggests that inter- 
strand crosslinks are unimportant. 12 ' 14 The site of platinum binding 
is most probably the guanos ine N(7) 15 > 16 and the platinum seems to be 
bound in a bidentate manner. 17 

Relating Structure to Activity 

Since the discovery cf the anti-tumor activity of cis-Pt (NH 3 ) 2 C1 2 
was empirical, if not accidental, we have no reason to believe that 
it is the best drug in its class. Efforts to design better drugs 
should be aided by "Rosenberg's Rules." 5 Several structure-activity 
relationship studies have been undertaken, ia > 19 but no conclusive 
results have been reported. 



-27- 

Conclusion 

Through the efforts of researchers in this field, more effec- 
tive compounds have been found ( e. g . , the "platinum-pyrimidine blue" 
complexes). 5 Further efforts will lead, one hopes, not only to yet 
more effective complexes, but to an understanding of the mode of 
action of these platinum species and even to a better understanding 
of the nature of cancer. 

References ' 

1. B. Rosenberg, L. Van Camp, J.E. Trosko, V.H. Mansour, Nature . 222 . 
385 (1969) 

2. R.C. Lange, R.P. Spencer, H.C. Harder, J. Nucl. Med ,. 14,191 (1973) 

3. E. Heinen, R. Bassleer, Biochem. Pharmacol .. 25 . 1871 (1976) 

4. S.K. Aggarwal, Cytobiologie . 8, 395 (1974) 

5. B. Rosenberg, Cancer Chemother. Rep. P. 1 . 59. 589 (1975) 

6. A. Khan, J.M. Hill, Transplantation . 13 . 55 (1972) 

7. F. Basolo, R.G. Pearson, "Mechanisms of Inorganic Reactions", 2 ed., 
Wiley, New York, 1967, p. 386 

8. M.C. Lim, R.B. Martin, J. Inorg. Nuc. Chem ., 38, 1911 (1976) 

9. A.J. Thomson, R.J. P. Williams, S. Reslova, Struct. Bonding (Berlin) . 
11,1 (1972) 

10. M.E. Friedman, J.E. Teggins, Bioch. Biophys. Acta . 350 . 263 (1974) 

11. B. Rosenberg, Die Naturwissenschaften . 60 . 399 (1973) 

12. L.L. Munchausen, R.O. Rahn, Cancer Chemother. Rep. P. 1 . 59 . 643 (1975) 

13. D.M.L. Goodgame, I.Jeeves, F.L.Phillips, A.C. Skapski, Bioch . 
Biophys. Acta , 378 . 153 (1975) 

14. H.C. Harder. Chem. -Biol. Interact . 10 . 27 (1975) 

15. P.C. Kong, T. Theophanides, Inorg. Chem . 13 . 1167 (1974) 

16. A.B. Robins, Chem.-Diol. Interact. . 6. 35 (1973) 

17. J. -P. Macguet, T. Theophanides, Bioinorg. Chem. , 5 . 59 (1975) 

18. M.J. Cleare, J.D. Hoeschele, Platinum Metals Rev .. 17. 2 (1973) 

19. P.D. Braddock, T.A. Connors, M. Jones, A.R. Khokhar, D.H. Melzak, 
M.L. Tobe, Chem. -Biol. Interact .. l_l r 145 (1975) 



-28- 



STUDIES OF SOME DYNAMIC METAL CARBONYL CLUSTER COMPLEXES 



Steve Richter 



December 7, 1976 



Interest in the study of relatively small, discrete polynuclear 
compounds containing metal-metal bonds, or clusters, is steadily in- 
creasing. This is in part due to the use of clusters as models for 
the structures and mobility of chemisorbed species on metal surfaces 
and for the elucidation of reaction mechanisms in heterogeneous 
catalysis. The dynamic behavior of three classes of ligands, namely 
hydrides, carbonyls, and hydrocarbon species, in a cluster has been 
examined in detail through the use of NMR. 



In 
for hydr 
hydrido 
synthesi 
one isom 
two stru 
structur 
diphos , 
3 of the 
based on 
were gen 
resembl e 
physical 
scheme . 



an attempt to distinguish between the proposed mechanisms 
ide scrambling, edge- terminal -edge and edge- face-edge , the 
cluster H4RU4CCO) io(Diphos) [Diphos = Ph2PCH 2 CH2PPh 2 ] was 
zed. The slow exchange *H NMR spectrum indicates that only 
er is present with four nonequival ent hydride sites. Of the 
ctures consistent with the observed P-H splittings, one 
e with a chelating diphos and the other with a bridging 
the latter is assumed to be correct. At room temperature, 
4 sites have been equilibrated. Computer simulated spectra 
all the possible permutations for averaging the three sites 
erated. Only simulations of three of these permutations closely 
the experimental spectra. They uniquely correspond to a 
process consisting of an edge- face- edge hydride scrambling 



13 
Through the use of C 

and hydrocarbon ligand mobili 

of the formula OS3 (CO) 1 q (dien 

temperature spectra are consi 

to one Os atom either in an u 

symmetrical "bicapped" fashio 

transferred from one Os to an 

conjugated dienes suggests th 

involves concerted restricted 

of the diene and two adjoinin 

intracluster CO scrambling oc 

intermediate, where 3 CO's ar 

In the case of noncon j ugated 

rearrangement process involve 

combination of mechanisms pro 



NMR, informatio 
ties in a serie 
e) has been obt 
stent with the 
nsymmetrical ax 
n. Although th 
other, the NMR 
at the initial 

or complete ps 
g carbonyls . A 
curs, presumabl 
e bridging the 
dienes, such as 
s a tribridging 
ceeding at high 



n concerni 
s of trios 
ained. Th 
diene unit 
ial-equato 
e diene li 
line shape 
site excha 
eudo- four 
t higher t 
y through 
edges of t 
norbornad 
intermedi 
er tempera 



ng carbonyl 
mium clusters 
e limiting low 

coordinated 
rial or a 
gand is not 

behavior for 
nge process 
fold rotation 
emperatures , 
a tribridging 
he OS3 triangle, 
iene, the initial 
ate, with a 
tures . 



Additional studies are being conducted to determine the generality 
of the above mechanisms for ligand mobility in clusters. 

General References 



1. E.L. Muetterties, Bull . Soc . Chim. B e 1 g . , 84 , 959(1975). 

2. H.D. Kaesz, Chem . Brit . , 9, 344(1973). 

3. L.M. Jackman and F.A. Cotton, Eds., "Dynamic Nuclear Magnetic 
Resonance," Academic Press, New York, N.Y., 1975. 



PHOTOCHEMICAL SUBSTITUTION OF MANGANESE DECACARBONYL 
Dennis Kidd December 9, 1976 

Since 1969, several mechanistic studies have been carried 
out on the Lewis base substitution of Mno(C0),Q. Several mechanisms 
for the substitution have been proposed. It is now generally agreed 
that the first step is rupture of the Mn-Mn bond to form the 17 
electron Mn(CO),-. The substitution of Mn(C0)r before it recombines 
to yield substituted dimer is not well understood. Therefore the 
photochemical reactions of Mn 2 (C0)-.Q with PBu^ or P(0Et) 3 have 
been studied. The Mn(C0) 5 radical was generated by irradiation 
of the absorption at 350 nm, corresponding to the o+o* transition 
in Mn 2 (C0),Q. The intensity of the light source was determined 
by K.,Fe-| ( C2CK) o actinometry. 

There are two likely mechanisms for the substitution of 
Mn(C0) 5 ; a 2nd order path involving a 19 electron intermediate, (1) 

Mn(C0) 5 + L ■+ [!_■ • -Mn(C0) 4 - • -Co] + Mn(C0) 4 L + CO (1) 

and a dissociative mechanism involving a 15 electron intermediate 
(2-3). 

Mn(C0) 5 * Mn(C0) 4 + CO (2) 

Mn(C0) 4 + L z Mn(C0) 4 L (3) 

The distribution of substituted dimers is different for PBu., 
and P(0Et) 3 . PBu 3 forms Mn 2 (C0)gL 2 and Mn(C0) 3 l_ 2 , while P(0Et) 3 
forms Mn 2 (C0)gL 2 , Mn 2 (C0) 7 l_ 3 , and Mn(C0) 3 L 2 . This difference is 
probably due to a difference in steric requirements of the ligands. 
However, the quantum yield and the pseudo first order rate constant 
for disappearance of Mn 2 (C0)-,Q are the same in both cases and are 
independent of the concentration of L. Saturating the solution 
with CO lowers both the quantum yield and the rate constant. 

The data are in best agreement with the dissociative mechanism. 
The results may be generalized to provide a working hypothesis: 
When a single electron occupies a stereochemical site in a metal 
carbonyl radical it induces lability. 



-30- 

HIGH PRESSURE EFFECTS ON TRANSITION METAL COMPLEXES 

Virgil L. Payne February 8, 1977 

Introduction , 

In much of scientific research, observation of the change in 
a property of a system with temperature is a common means of study- 
ing the system. Within the past twenty years, a substantial effort 
has been made in the development, refinement and application of high 
pressure (>1 kbar=987 atmospheres) techniques to the study of physi- 
cal and chemicJal behavior. These techniques have been responsible 
for the synthesis of new materials, the study of kinetics and mechan- 
isms of reactions and the structural and electronic characterization 
of compounds. 1 * 2 

Experimental Apparatus 

Much of the development of experimental high pressure apparatus 3 * 4 
is due to the pioneering work of P. W. Bridgeman in the 19^0' s and 50' s. 
Utilizing what he termed the "principle of massive support," Bridgeman 
constructed a simple anvil device for electrical resistance measure- 
ments which would generate pressures to 50 kbars. Drickamer has modi- 
fied the Bridgeman device by tapering the anvils to create a radially 
decreasing pressure gradient. When using NaCl as the pressure- 
transmitting substance, this cell can be used with optical techniques 
to pressures of 200 kbars. Another modification of the Bridgeman de- 
vice' used with optical techniques is that of the Diamond Anvil Cell 
(DAC). 5 The anvils are Type II diamonds which absorb radiation at 
^> 2000 cm" 1 , but are transparent below 1000 cm" 1 and can generate 
pressures in excess of 350 kbars. Various modifications of these and 
other devices have been made for low and high temperature work. 

Calibration of high pressure devices has been based on distinct 
changes in electrical resistance of such materials as Iron, Barium or 
Bismuth. 3 * 6 In recent years, the optical devices have been calibrated 
using methods based on the pressure-dependent changes in the fluores- 
cence spectrum of such materials as ruby. 7 

Applications 

Ferraro and coworkers 8 * 9 have examined several transition metal 
complexes in order to observe structural nonrigidity in the solid state, 
They have observed conversion of tetrahedral Ni(BzPh 2 P) 2Br 2 to its 
square planar isomer and also trigonal-bipyramidal Ni(CN) 5 ~ 3 to its 
square-pyramidal isomer. In both cases, the complexes exist as a 
mixture of isomers at ambient pressure with conversion to only one 
isomer occurring upon application of sufficient pressure. 

Six-coordinate complexes have not been found to undergo struc- 
tural changes, but a few undergo interesting electronic changes. 
Drickamer and coworkers 10 have observed both high spin to low spin 
and low spin to high spin state changes in bis- and tris(phenan- 
throllne) iron(ll) complexes as well as reduction of Fe(lll) to Fe(ll) 
in B-diketone complexes. These findings are interpreted as increased 
covalency in the metal- ligand bond, decreased n-backbonding and vari- 
ation of electronic properties of the ligand with pressure. 



-31- 

Poorly conducting one- dimensional complexes such as metal 
glyoximes 11 and MPt(CN) 4 (M=Ba 2 , Sr 2 , Mg 2 ) 12 experience a dramatic 
increase in conductivity with pressure, although metallic behavior 
cannot be achieved using nondestructive methods. Similar behavior 
has also been observed in mixed-valence complexes of Platinum and 
Palladium. 13 * 14 Variation of temperature and pressure has revealed 
evidence for a sharp metal-to- semiconductor transition in 
K 2 Pt(CN)4Bro.3o(H 2 0)3 (KCP). 15 This is interpreted in terms of in- 
creased interchain coupling which would suppress the Peierl's dis- 
tortion. Superconductivity of KCP under pressures > 70 kbars is a 
possible consequence of this behavior. 

Conclusion 

The further development of high pressure techniques capable of 
extending the range of pressure should continue to provide chemists 
and physicists with a better understanding of the structure, physical 
properties and behavior of materials. 

References 

1. J. R. Ferraro and G. J. Long, Acc'ts Chem. Res., 8, 171 (1975). 

2. R. Sinn, Coord. Chem. Rev., 12, 185 (1974). 

3. C. C. Bradley, "High Pressure Methods in Solid State Research," 
Plenum Press, N.Y. , 1969. 

k. R. H. Wentorf, "Modern Very High Pressure Techniques," 
Butterworths, London, 1962. 

5. S. Block and G. Piermarini, Physics Today, 29, 44 (1976). 

6. H. G. Drickamer, Rev. Sci. Inst., 4l, 1667X^971). 

7. J. D. Barnet, S. Block and G. Piermarini, Rev. Sci. Inst., 44, 
1 (1973). 

8. J. R. Ferraro, et al. , Inorg. Chem., 13, 496 (1974). 

9. J. R. Ferraro, et al., J. Chem. Soc, Chem. Commun. , 17, 
827 (1975). ~ 

10. H. G. Drickamer and C. W. Frank, "Electronic Transitions and 
the High Pressure Physics and Chemistry of Solids," Chapman and 
Hall, London, 1973. 

11. Y. Kara and I. Shirotani, Solid State Commun., 19, 171 (1976). 

12. Y. Hara and I. Shirotani, ibid., 17* 827 (1975). 

13. L. V. Interrante and F. P. Bundy, Inorg. Chem., 10, II69 (1971). 

14. L. V. Interrante, K. ¥. Browall and F. P. Bundy, Inorg. Chem., 
13, 1158 (1974). 

15. M. Thielemans, et al. , Solid State Commun., 19, 21 (1976). 



RADICAL CHAIN PATHWAYS IN SUBSTITUTION REACTIONS OF METAL CARBONYL 

COMPOUNDS 

Ma'mum Absi-Halabi (Final Seminar) March 1, 1977 

Metal carbonyl compounds have long been known to undergo substi- 
tution by either dissociative, associative, dissociative interchange 
or ligand migration mechanisms. However, recent work on some carbonyl 
metal hydrides and manganese carbonyl compounds has revealed evidence 
of novel substitution pathways 1 proceeding by carbonyl radical inter- 
mediates. * 

The reactions of Cl 3 SnCo(C0) u with some group V bases, resulting 
in the formation of the ionic compounds, [Co (CO) 3 B 2 ] [SnCl 3 ] , have 
been investigated under various conditions. It has been observed 
that oxygen, galvinoxyl, tetracyanoethylene and trichlorobromomethane 
inhibit the reactions. On the ether hand, visible radiation from a 
fluorescent lamp or one electron donor compounds such as anhydrous 
Cr (acac ) 2 promote the reactions. The order of the reaction with re- 
spect to the metal carbonyl is 3/2, and the estimated ratios of the 
observed rate constants for AsPh 3 , PPh 3 and PBu 3 are approximately 
1:150:16000. 

These results suggest that the reactions proceed by a free-radical 
chain mechanism. The chain is initiated by homolytic cleavage of the 
.metal-metal bond, resulting in formation of a Co(C0) 4 » radical which 
undergoes rapid substitution with the base. An important step in 
the mechanism is an outer-sphere electron transfer process from a 
Co(C0) 3 B« radical to a Cl 3 SnCo(C0) u molecule: 

Co(C0) 3 B» + Cl 3 SnCo(C0) u •> Co(C0) 3 B + + Cl 3 SnCo ( CO ) u •" 

Although our observations are new to the field of mechanistic 
studies of metal carbonyls, similar ones have been reported 2 in organic 
chemistry. Furthermore, numerous substitution reactions of metal car- 
bonyls, such as V(C0) 6 , Co 2 (C0) B and Mn 2 (C0)i O , apparently proceed 
by pathways analogous to the one we have proposed. This is verified 
by results obtained for the dicobalt octacarbonyl reactions with PPh 3 
and PBu 3 . 

Our results add further evidence for the importance of carbonyl 
radicals in substitution reactions of metal carbonyls. Moreover, a 
new phenomenon in metal carbonyl mechanistic studies has been estab- 
lished, namely, electron transfer processes promoted by base substi - 
'-tut ion. 

References 

1. a. B. H. Byers and T. L. Brown, J. Am. Chem. Soc, _97, 3260 (1975) 

b. D. G. DeWit, J. P. Fawcett and A. J. Poe , J. Chem. Soc. Dalton, 
528 (1976). 

c. D. R. Kidd and T. L. Brown, in preparation. 

2. a. N. Kornblum, Angew. Chem. Internat. Edit., l£, 73^ (1975). 
b. J. Pinson and J. M. Saveant, J. Chem. Soc. Chem. Comm. , 

933 (197^). 



* -33- 

ELECTROCHEMICAL METHODS IN BIOINORGANIC CHEMISTRY 

Steven L. Suib March 24, 1977 

Introduction 

The science of electrochemistry during the last 20 years has 
gone through a period of great growth and change. 1- ^ Recently a 
substantial effort has been made to develop electrochemical tech- 
niques that are particularly applicable to the fields of bioelec- 
trochemistry and bioenergetics . Several model systems have been 
studied to obtain information concerning analytical measures of 
concentration and the kinetics and mechanisms of bioinorganic 
reactions . 

Techniques 

Pulse polarographic techniques have been developed mainly 
to obtain increased sensitivity and resolution for analytical 
measurements of concentration. >9 The improvements manifested 
in differential pulse polarography (DPP) are about three orders 
of magnitude better than those of d.c. polarography. In addition, 
information concerning specific oxidation states and complexation 
can be obtained, however, pulse techniques are rarely used for 
diagnosis of electrode processes. 

Cyclic voltammetry has been widely used to investigate the 
reversibility of electron transfer reactions .- L0 ' H It is now a 
powerful technique for studying irreversible electron transfer 
processes, electron transfer reactions preceded or followed by 
relatively slow chemical reactions, and adsorption effects at 
electrodes. In addition, the results of theoretical calcula- 
tions have made it possible to use cyclic voltammetry to 
measure standard rate constants. 

Kuwana, Heineman and Winograd have been the main workers 
in spectroelectrochemistry , which entails the use of an opti- 
cally transparent electrode (OTE). 12-1 ^ The main utility of 
the OTE is in elucidating electrode reactions and determining 
stoichiometry , energetics, molar absorptivities and diffusion 
coefficients of electrogenerated intermediates or products. 

Applications 

Holm has recently studied iron-sulfur proteins using d.c. 
polarography, cyclic voltammetry and differential pulse polaro- 
graphy. 15» Id Using differential pulse polarography, a mixture 
of the 2Fe-2S* dimer (Fe 2 S 2 (SPh) u )~ 2 and the 4Fe-4S* cluster 
(Fe^Si* (SPh) n ) 2 has been electrochemically assayed. 1 ^ 

Correlations of kinetics and experimental parameters have 
made it possible to diagnose criteria so that unknown systems 
can be characterized. The variations of peak current, half- 
peak potential or ratio of anodic to cathodic peak currents 
are studied as a function of rate of voltage scan.^ Typical 
bioinorganic systems exploiting these diagnostic criteria are 



-34- 

featured in the work of Bard 1 ', Holm 1 "^ an d Pickett.l° Jordan 
and coworkers 1 -? have determined rate constants for various cyto- 
chrome c models using the method described by Nicholson and 
Shain. 11 

Spectrocoulometry with an optically transparent electrode 
cell has provided information about the bioelectrochemistry of 
cytochrome c and cytochrome c oxidase. 12 Since heterogeneous 
electron transfer from these hemes to the electrode is slow, redox 
titrations have been performed by the coulometric generation of 
mediator titrants. The specific mechanisms of the electrode 
reactions and the rate constants for this system have been ana- 
lyzed. 

Summary 

Differential pulse polarography , cyclic voltammetry and 
spectroelectrochemistry are only a few of the techniques that 
can provide simple, rapid and reliable results for the bioinor- 
ganic chemist. The main offerings of electrochemistry will 
undoubtedly be in the realm of electrode mechanisms and asso- 
ciated kinetics. In time, more chemists, biologists, physio- 
logists and clinicians will work hand-in-hand with the electro- 
chemist to understand systems which cannot be unraveled by the 
usual chemical methods . 

References 

1. F. Anson, "Electroanalytical Chemistry," An ACS Audio 
Course (1976). 

2. R. N. Adams, "Electrochemistry At Solid Electrodes," 
Marcel Dekker, Inc., N.Y., 1969. 

3. D. R. Crow, "Principles and Applications of Electrochem- 
istry," Chapman and Hall, London, 1974. 

4. J. B. Keadridge, "Electrochemical Techniques for Inorganic 
Chemists," Academic Press, N.Y., 1969- 

5. W. C. Purdy, "Electroanalytical Methods in Biochemistry," 
McGraw-Hill, N.Y., 1965- 

6. D. T. Sawyer and J. L. Roberts, Jr., "Experimental Electro- 
chemistry for Chemists," John Wiley & Sons, Inc., N.Y., 1974. 

7. D. Pletcher, Chem. Soc . Rev., ±, 4'fl (1975). 

8. D. E. Burge, J. Chem. Ed., 4_7, A8l (1970). 

9. J. Osteryoung and R. A. Osteryoung, Amer. Lab., July (1972). 

10. R. S. Nicholson and I. Shain, Anal. Chem., 36, 706 (1964). 

11. R. S. Nicholson, Anal. Chem., 37., 1351 (1965). 

12. T. Kuwana and W. R. Heineman, Acc'ts Chem. Res., 9, 24l (1976) 

13. T. Kuwana and W. R. Heineman, Bioelectrochem. and Bioener- 
getics, 1, 389 (1974). 

14. A. J. Bard, Ed., "Electroanalytical Chemistry," Marcel Dekker, 
Inc. , N.Y. , 1974. 

15. R. H. Holm, et al., J. Am. Chem. Soc, 96, 4159 (1974). 

16. R. H. Holm, et al., J. Am. Chem. Soc, 97, 1032 (1975). 

17. A. J. Bard, et al . , J. Am. Chem. Soc, <gf» ^ 872 (1972). 

18. C. J. Pickett, et al . , J. C. S. Chem. Comm. , 22, 94l (1976). 

19. J- Jordan, et al., Bioelectrochem. and Bioenergetics , 1, 73 
(1974). 



-35- 

ELECTRON TRANSPORT IN CHLOROPHYLL 

John R. Long March 29, 1977 

The role(s) of the numerous types of chlorophylls in photo- 
synthesis are of contemporary research interest. It has been found 
that photosynthetic plants and bacteria consist mainly of antenna 
chlorophyll molecules: 1 » 2 > 3 they, along with carotenoids, serve to 
collect the light energy and funnel it to a "complex" that converts 
it to a useful form of chemical energy. Such a complex is referred 
to as the "reaction center," symbolized as P, (X designates the 
principle absorption of the complex, 700 nm in plants, and 870 nm 
in bacteria) . 4 * 5 

CHLOROPHYLL DIMER 

As a result of this absorption of energy by the reaction center, 
there are several changes in the near IR region of the absorption 
spectra. The bands that change are believed to originate from bac- 
terio chlorophyll molecules (BChl, indicating that the chlorophyll 
originated from bacteria); this implicates BChl with the reaction 
center. Similar changes occur upon mild chemical oxidation of the 
samples, implying that normal photosynthesis involves oxidation of a 
BChl complex. 4 ' 5 ' 6 -» 7 Irradiation of jin vivo samples with red lig ht 
yields an ESR active species having a line width equal to 1/ V^T" 
that of in vitro monomeric BChl. 2 ENDOR studies have shown that the 
hyperf ine coupling constant of the in vivo BChl is 1/2 that of 
in vitro BChl. 15 These results are~Dest interpreted in terms of a 
dimeric BChl complex. If the reaction center is held at a low redox 
potential, -100 mv, triplet ESR spectra exhibiting ESP and reduced 
zero field splitting parameters (relative to in vitro monomeric BChl) 
are observed. These observations are best interpreted in terms of 
delcalization of the electrons over more than one BChl molecule. 8 " 14 
The optical spectra are also consistent with a dimeric structure. x * 2 > L1 

REACTION MECHANISM 

The primary photochemical reaction is considered to be the 
transfer of an electron from an excited "BChl special pair," P*, 
to some primary electron acceptor, X, such that P* and X* are formed. 4 " 6 
If the redox potential is made low enough, ^j -100 mv, such that X is 
prereduced, the optical spectra show the presence of a s,hort 
(t^IO ns) and a long lived (t = 6 jus) state, labeled P and P R 
respectively. 3 > Q > 16 , 9 , 17 Analysis of the kinetics, down to the pico- 
second time scale, of these spates reveals that P* forms directly 
from P* 6 and that the state P itself involves some form of charge 
or electron transfer. e > 9 Other optical studies have shown that P 
also forms as an intermediate in the normal (ie., X is not prereduced) 
primary photochemical process. 16 These two pieces of information 
indicate that there must be an intermediate electron acceptor, 
symbolized I. ESR and optical data indicate that this species^is 
most likely a bacteriopheophytin (BPh). 9 * 19 At the present, P is 
the only observable intermediate in the normal photo-oxidation of P. 



-36- 

•p "p 

The other transient state, P , forms directly from P and only forms 

under conditions where X is prereduced^ 5 P is believed to be the 
ESR observed triplet. The nature of P is quite controversial, 
speculations range from charge transfer within the BChl reaction 
center: [BChl --BChl" ], to simply an oxidized BChl dimer: [BChl — - 
BChl] . 5 * *> B ' 10, 1B ' 19 Steady state ESR studies of reaction centers 
have shown a signal which is attributed to the primary electron 
acceptor, X, which appears to be a ubiquinone molecule (UBQ). Pico- 
second studies have also shown that UBQ acts as the primary electron 
acceptor. 2 °> 21 

A large volume of experimental data has been amassed, yet the 
same data is still subject to numerous interpretations. Present 
and future work lies in unambiguously identifying the exact character 
of the different species/states that are formed. 22 

References 

1. T. M. Cotton, A. D. Trifunac, K. Ballschimiter and J. J. Katz, 

Bioc. Biop. Acta, 368, l8l (197*0. 
.2. J. R. Norris, R. A. Uphaus, H. L. Crepsi, and J. J. Katz, 
Proc. Nat. Acad. Sci. U.S.A., 68, 625 (1971). 

3. J. Franck,and J. L. Rosenberg, J. Theoret. Biol., 7, 276 (1964). 

4. W. W. Parson and R. J. Cogdell, Bioc. Biop. Acta, 4l6, 105 (1975) 

5. W. W. Parson, R. K. Clayton and R. J. Cogdell, Bioc. Biop. Acta, 
387, 265 (1975). 

6. M. G. Rockley, M. W. Windsor, R. J. Cogdell and W. W. Parson, 
Proc. Nat. Acad. Sci. U.S.A., 72, 2251 (1975). 

7. D. C. Borg, J. Fajer, R. H. Felton and D. Dolphin, 
Proc. Nat. Acad. Sci. U.S.A., 67, 813 (1970). 

8. J. J. Leigh and P. L. Dutton, Bioc. Biop. Acta, 357, 67 (197*0. 

9. P. L. Dutton, K. J. Kaufmann, B. Chance, P. M. Rentzepis, 
FEBS Lett., Vol. 60 (2), 275 (1975). 

10. R. A. Uphaus, J. R. Norris and J. J. Katz, Bioc. Biop. Res. 
Commun. , 6l, 1057 (1974). 

11. J. R. Norris, R. A. Uphaus and J. J. Katz, Chem. P. Lett., 

31(1) :157 (75). 

12. M. C. Thurnauer, J. J. Katz and J. R. Norris, Proc. Nat. Acad. 
Sci. U.S.A., 72, 327 (1975). 

13. R. H. Clarke and R. E. Connors, Chem. P. Lett., 42, 69 (1976). 

14. J. F. Kleibeuk, R. J. Platenkamp and T. J. Schaafsma, 
Chem. P. Lett., 4l, 557 (1976). 

15. J. R. Norris, H. Scheer, M. E. Druyan and J. J. Katz, 
Chem. P. Lett., 41, 557 (1976). 

16. K. J. Kaufmann, P. L. Dutton, T. L. Netzel, J. S. Leigh and 
P. M. Rentzipis, Sci., 188, 1301 (1975). 

17. R. T. Cogdell, T. G. Monger and W. W. Parson, Bioc. Biop. Acta, 
408, 189 (1975). 

18. K. J. Kaufmann, K. M. Petty, P. L. Dutton and P. M. Rentzepis, 
Bioc. Biop. R., 70, 839 (1976). 

19. J. Fajer, D. C. Brune, M. S. Davis, A. Forman and 

L. D. Spaulding, Proc. Nat. Acad. Sci. U.S.A., 72, 4956 (1975). 



-37- 

20. A. R. Mcintosh and J. R. Bolton, Bloc. Biop. Acta, 430, 
555 (1976). 

21. G. Feher, M. Y. Okamura and J. D. McElroy, Bioc. Biop. Acta, 
267, 222 (1972). 

22. J. A. Anton, J. Kwong and P. A. Loach, J. Hetero. Ch. , 13, 
717 (1976). 



-38- 



STRUCTURAL AND ELECTRONIC PROPERTIES OF 
BIS(CYCLOPENTADIENYL) TITANIUM AND VANADIUM COMPOUNDS 



Benjamin Fieselmann 



(Final Seminar) 



March 31, 1977 



The study of low valent early transition metal compounds has 
been of increasing interest because of their ability to catalyze 
reactions such as the reduction of nitrogen to ammonia and the 
isomerization of olefins. 1 * 2 Although reduced forms of titanium 
and vanadium* tend to be reactive, especially with air, we have syn- 
thesized several thermally stable monomeric, dimeric, and trimeric 
low valent compounds which allowed us to study steric and electronic 
features and to better define the requirements for thermal stability. 

Two different dimeric compounds were prepared by complexing two 
bis ( eye lopentadienyl) titanium (ill) units each to 2, 2'biimidazole(l) 
and 2,2'bibenzimidazole(2) . The variable temperature magnetic sus- 
ceptibility of these two compounds showed antiferromagnetic coupling 
(j = -25.2 cm" 1 and -19.2 cm" 1 respectively). The small observed 
coupling here and in later compounds is fully consistent with the 
Alcock model and more recent ESR studies that place the d 1 electron 
in a largely nonbonding orbital between the cyclopentadiene rings. 3 
The ESR spectra for both dimers in the glass phase showed an S=l 
triplet state spectra that can only be explained by the coupling of 



CrCp 



1 Cpcp 




Cp Cp 




N CH 



two adjacent spins. 4 The monomer bis(cyclopentadienyl)titanium(lIl) 
met hylbenz imidazole (3) was prepared and the ESR in the glass showed, 
in addition to titanium hyperfine (A = 11. G), nitrogen hyperfine 
(A = 2.2G), which indicates derealization of charge onto the bridge 
and suggests an exchange pathway through the bridge. The crystal 
structure of the titanium dimer of biimidazole confirms the dimeric 
structure and suggests that the chelate effect of the rigid bidentate 
bridge contributes to the thermal stability of the compounds. 

A similar dimer was prepared where two bis (eye lopentadienyl) - 
titanium(lll) groups are bridged by two pyrazole rings(4). Although 
the ESR spectrum is a typical S = 1 spectrum generated from inter- 
acting metal centers, the magnetic susceptibility shows no coupling 
down to the temperature of liquid helium. This lack of coupling 
could result from too small a J value to observe because of incorrect 
ligand orbital symmetry or from crystal packing forces in the solid 
which disrupt the coupling pathway. Single crystal X-ray structural 
studies indicate that the configuration of the dimer is that given 
below (4). 

V 





-39- 

In 1976, E. 0. Fischer 5 reported one of the few stable organo- 
metallic t Itanium ( II ) compounds, bis(cyclopentadienyl)titanium(Il) 
bipyridyl(5) . In addition to its thermal stability, this d 2 compound 
is of interest because it has an anomalous magnetic moment of about 
.6 B.M. The magnetic moment can be explained successfully by showing 
that the molecule has a ground state singlet and an occupied excited 
triplet state. The solid state ESR intensity and the solution mag- 
netic susceptibility determined by NMR (Evans Method) 6 were both com- 
puter fit to theoretical expressions for singlet-triplet behavior 
that allowed 'calculation of the energy separation (E = 758 cm" 1 in 
the solid state, 592 cm" 1 in solution). 7 The ESR in a glass shows 
a triplet spectrum indicating coupling between the electrons. The 
structure of the triplet state can best be described as a Ti(lll) and 
a bipyridyl anion. 

The structure of the isoelectronic d 2 compound bis(cyclopenta- 
dienyl) vanadium (ill) monochloride was undertaken to provide further 
steric and electronic information on d 2 compounds. Unlike the dimeric 
monochlorides of bis(cyclopentadienyl) scandium and titanium, the 
vanadium compound is monomeric with the chlorine atom residing sym- 
metrically in the open face of the tilted cyclopentadiene rings. 
The monomeric structure and paramagnetism (2.80.B.M. ) can be explained 
by examining the molecular orbitals of the Cp 2 V moiety. 8 A degener- 
acy between two orbitals results in two unpaired electrons. The sym- 
metric structure results both from steric considerations and from 
electronic factors which attempt to minimize electronic repulsion 
between the ligand and metal. 

C P pP 

cr T \ ; 




>~N 
7& 



Two trimers of cyanuric acid with three Cp 2 Ti(lIl) and three 
(MeCp) 2 Ti(lIl) units have been synthesized (6). The trimers are 
thermally stable at over 250°C and have very intense molecular ion 
peaks in the mass spectra. The Ti(lll) has a large enough atomic 
radius to form a four member ring and chelate with the oxygen and 
nitrogen atoms. The ESR spectrum in a glass can be interpreted as 
arising from both the populated quartet state and the doublet states. 9 
The magnetic susceptibility shows antiferromagnetic coupling and 
can be fit to a J value of -.9^ cm" 1 . The monomer 
bis ( eye lopentadienyl) titanium (ill) 2-hydroxypyridine (7) shows and 
ESR signal unlike that of the trimer, which confirms that even with a 
very small coupling, the trimer' s electronic structure is more 
complicated than just the sum of the properties of three independent 
monomers. The most likely pathway of exchange is a superexchange route 
through either the nitrogen or oxygen or both of the cyanuric acid. 



-40- 



References 



1. R. D. Sanner, D. M. Duggan, T. C. McKenzie, R. E. Marsh and 

J. E. Bercaw, J. Am. Chem, Soc , 98, 8358 (1976). 

2. G. P. Fez and S. C. Kwan, J. Am. Chem. Soc , 98, 8079 (1976). 

3. J. L. Petersen and L. F. Dahl, J. Am. Chem. Soc . , 97, 64l6 

(1975) ,and J. Am. Chem. Soc ., 97, 6422 (1975). 

4. E. Wassermann, L. C. Snyder and W. A. Yager, J. Chem. Phys ., 

41, 1763 (1965); J. E. Wertz and J. R. Bolton, 
Electron Spin Resonance , McGraw-Hill, 1972, p. 238. 

5. E. 0. Fischer and R. Amtmann, J. Organometal. Chem. , 9, 15 (1967). 

6. B. D. F. Evan, J. Chem. Soc , 2003 (1959); I. B. Joedicke, 

H. V. Studer and J. T. York, Inorg. Chem ., 15, 1352 (1976). 

7. D. Bijl, H. Kainer and A. C. Rose-Innes, J. Chem. Phys ., 30, 

765 (1959). 

8. J. W. Lauher and R. Hoffmann, J. Am. Chem. Soc , 98, 1929 (1976). 

9. J. Brickmann and G. Kothe, J. Chem. Phys ., 59, 2807 (1973). 



-40- 



EXPERIMENTALLY DETERMINED RATES OF THERMAL ELECTRON 
TRANSFER IN MIXED- VALENCE IRON COMPOUNDS 

Ronald G. Wollmann (Final Thesis) April 7, 1977 

A mixed- vale nee compound typically contains two or more trans- 
ition metal ions that are in different formal oxidation states. 1 * 2 
An electron(s) may, thus, be thermally transferred between transition 
metal ions. ' Experimental determinations of the rates of thermal 
electron transfer and development of a thorough understanding of the 
effects of chemical environments and changes in chemical environ- 
ments on these rates are of interest. The insight gained from the 
investigation of simple mixed- valence compounds may be pertinent to 
understanding electron transfer processes in the biologically im- 
portant class of compounds known as electron-transfer proteins. 3 

One type of mixed- valence compound investigated in this work is 
derived from u-oxo-bis(tetraphenylporphinatoiron(lIl) ) , [Fe(TPP)] 2 0, 
and u-oxo-bis (N,N' -ethylenebis (salicylideneiminatoiron(lll) ) , 
[Fe(salen) ] 2 0. These compounds are chemically oxidized to the corre- 
sponding oxo-bridged diiron(lII,Iv) compounds, [Fe 2 (TPP) 2 0]X and 
[Fe 2 (salen) 2 0]X, where X" is variously PF 6 ", BF 4 ~, C10 4 ~ or I 3 ". 
Antiferromagnetic exchange interactions are indicated in the variable 
temperature (4.2->267 k) magnetic susceptibility data which are least- 
squares fit to the theoretical equation for an isotropic exchange 
interaction (h = -2JS 2 »S 2 ) in an S x = 5/2, S 2 = 2 dimer. The exchange 
parameters, J, for the [Fe 2 (salen) 2 0]X compounds are in the range of 
-7.5 to -17.6 cm" 1 . Inclusion of an axial zero-field interaction, 
DS 2 , in the theoretical susceptibility equation for the [Fe 2 (TPP) 2 0]X 
compounds yields J values in the range of -83 to -119 cm" 1 and |D|- 
values in the range of 12 to 20 cm" 1 . 

A single, temperature-independent, quadrupole-split doublet is 
observed in the Mbssbauer spectrum of each compound. No inter- 
valence transfer band can be identified in the room temperature so- 
lution electronic absorption spectra of the [Fe 2 (TPP) 2 0]X compounds. 
Absorptions are observed at 450-500 nm in the electronic absorption 
spectra of the [Fe 2 (salen) 2 0]X compounds which are tentatively as- 
signed as intervalence transfer bands. Isotropic EPR signals having 
g-values of ~2.0 are observed for the [Fe 2 (salen) 2 ]X compounds. 
The signals have temperature-dependent line widths. Either a high- 
spin Fe(lll) or no EPR signal is observed for the [Fe 2 (TPP) 2 0]X 
compounds. These data indicate that the rates of thermal electron 
transfer are on the order of a^io 10 sec" 1 for these oxo-bridged 
diiron(lII,Iv) compounds. 

Magnetic exchange interactions and electron transfer through 
the conjugated system of a porphyrin ring have been investigated 
by the synthesis of a four-iron containing compound, meso-tetraferro- 
cenylporphyrin, H 2 TFcP, its copper complex, Cu(TFcP), and the corre- 
sponding oxidation products, [H 2 TFcP](x) 3 and [Cu(TFcP) ] (x) 2 where 
X is variously I 3 ~ or DDQH", the hydroquinone anion of 2,3- 
dichloro-5*6-dicyano-l,4-benzoquinone. No magnetic exchange 



-41- 



interactions are indicated in the variable-temperature (4.2->270 K) 
magnetic susceptibility data for the [H 2 TFcP](x)3 and [Cu(TFcP) ] (x) 2 
compounds. Typical ferricenium magnetic susceptibility data are 
obtained. No intervalence transfer bands can be identified in the 
electronic absorption spectra and EPR signals typical of ferricenium 
ions are observed. 

Variable -temperature (4,2>300 K) Mossbauer spectra of the 
[H 2 TFcP](x) 3 and [Cu(TFcP) j(x) 2 compounds consist of two quadrupole- 
split doublets that correspond to the ferrocene and ferricenium 
centers. Relative area ratios of the two doublets are temperature- 
dependent. As the temperature decreases, the ferrocene to ferri- 
cenium area ratio increases. The loss of the ferricenium center 
results from unpaired electron density migration from the ferri- 
cenium centers to the porpyrin ring. Unpaired electron density 
migration is postulated to occur when the dihedral angle between 
the plane of the porphyrin ring and the substituted cyclopenta- 
dienyl ring of the ferricenium center approaches 0°. This allows 
the n-systems of the two rings to strongly interact. The tempera- 
ture dependence is, possibly, the result of temperature dependent 
changes of the dihedral angle caused by crystal lattice packing 
rearrangements which are induced by counterions in the lattice. 

References : 

1. G. C. Allen and N. S. Hush, Prog. Inorg. Chem. , 8, 357 (1967); 
N. S. Hush, Prog. Inorg. Chem., 8, 391 (1967). 

2. M. B. Robin and P. Day, Adv. Inorg. Chem. Radiochem., 10, 
247 (1967). — 

3. G. R. Moore and R. J. P. Williams, Coord. Chem. Rev., 18, 
125 (1976). 



-42- 

THE ADSORBENT AND THE ADSORBATE: APPLICATION OF EPR TO CATALYSIS 

Muin S. Haddad April 12, 1977 

Introduction 

Since free radical intermediates are postulated in many cata- 
lytic mechanisms 1 and since many of the commercially important 
catalysts are supported transition metal ions, 2 the technique of 
electron paramagnetic resonance should be a useful tool in the study 
of catalytic 'processes. Due to its high sensitivity, it provides a 
unique technique for studying low concentrations of adsorbed inter- 
mediates and active sites. Epr spectroscopy is useful in unraveling 
the important aspects of catalysis on surfaces; the nature of the 
adsorbate and adsorbent and possible intermediates and mechanisms 
involved in the catalytic process. Several reviews have appeared in 
the literature. 3 " 7 

Application 

Transition Metal Ions . The chromia/silica system is an effective 
catalyst for polymerization of ethylene. It is believed that Cr 5 
is the active ion in the catalytic process. Epr 8 * 9 was used to study 
the nature of the active species. Cr 5 in a tetrahedral environment 
acts as a precursor to the active species which is thought to be a 
square-pyramidal complex that does not show an epr signal due to 
strong spin- lattice relaxation. 

Nicula and coworkers 10 studied the epr of Cu + ions in CuY zeo- 
lites. The two main objectives were to study the effect of the 
number of Cu ions per unit cell and the level of dehydration on the 
epr signal. Systems with four Cu ions per unit cell showed an 
isotropic single line at room temperature and an anisotropic signal 
with hyperfine structure at high temperatures. The authors explain 
their results in terms of a tumbling effect of Cu ions in the fully 
hydrated zeolites versus a limited tumbling in the partially de- 
hydrated systems. 

Migration of metal ions in zeolites is a very interesting 
property that might be important catalytically. Using epr, it was 
possible to probe the migration of Cu ions in CuMg, CuZn, and CuLa 
exchanged Y-type zeolites. 11 

Formation of fully coordinated complexes within the large 
eavities of zeolites where the framework oxygens need not be in the 
first coordination sphere of the metal ion provides a new role for 
zeolites as coordination catalysts. Lunsford and Vansant 12 were 
able to stabilize penta- and hexacoordinate Co ( II) -methyl isocyanide 
in Co(ll)caY zeolites. Line shape of the epr signal, hyperfine due 
to 59 Co (l=£ ), and superhyperfine due to 13 C in CH 3 N 13 C were used to 
detect the nature of the species and its ground state. 



-43- 

The reversible uptake of oxygen by Co(ll) complexes in solution 
has been extensively studied, particularly with a view to under- 
standing the bonding of oxygen in biological oxygen carriers. Pro- 
longed exposure of ammoniated CoY zeolite to molecular oxygen results 
in an epr signal that shows 15- line hyperfine structure due to the 
interaction of the unpaired electron with two cobalt centers. 13 The 
signal was attributed to u-superoxo Co (III) dimer cation: 

[(NHaJsCo^O^Co 111 ] 5 ' - . 



Adsorbed Species . The most studied adsorbed species is_oxygen and 
the radicals formed from it on surfaces, namely, 2 ~, Q~ and 3 ~. 
Conclusive evidence for the formation of 2 ~ was obtained by observing 
the epr signal of an 17 0(l=5) enriched sample on Y-i rra diated MgO. 14 
An eleven- line hyperfine pattern was seen in the spectrum as expected 
for a diatomic species with both labeled oxygen atoms. The equiva- 
lent 17 hyperfine value of 77G indicates that 2 ~ is interacting with 
the surface in a side-on fashion. A peroxy-type of bonding of 2 ~ 
was detected by the epr of 17 enriched 2 on a decationated zeolite 
and Mo(VT) supported on silica gel. 15 Two sets of six equally spaced 
hyperfine lines were observed, indicating that 2 ~ with one labeled 
oxygen is formed where, in one case, 17 is bonded to the surface 
and, in another, it is not. 

Lunsford and Wang 16 observed a 6- line hyperfine pattern in tljie 
spectrum of 2 ~ on Y-irradiated NH 4 Y zeolite and concluded Al (1=5) 
is the Lewis acid site and there is only one type of Al interacting 
with 2 ~ on the surface. 

By adsorbing 13 C0 on Thoria 17 two types of CO adsorbates were 
detected by the observation of two sets of four hyperfine lines 
centered around g lf . From spin density calculations and conductivity 
measurements, it was concluded that chemisorption of CO results in 
the formation of slightly positive CO adsorbates and surface becoming 
slightly negative. 

Contact of CO with Ti0 2 with pre-adsorbed oxygen results in the 
formation of the species [0 I -0 TT -C0- rTT ] whose identification was 

made feasible by observing hyperfine patterns due to 13 C and 17 in 
an enriched sample. 18 The magnitude of the 17 hyperfine coupling 
constant increases in going from TTT to T . This is explained in 
terms of the species bonded to the surface at TTT pulling most of 
the negative charge towards the surface. 

Recently, 19 sulfur dioxide anions were formed in NH4Y zeolites 
at 200°C. The analysis of the line shape of the epr signal as a 
function of temperature * time at constant temperature, and adsorbed 
2 pressure led to proposing that S0 2 ~ occurs at two sites, A and B, 
in the zeolite. Site A is located in the supercage and Site B might 
be located in the sodalite unit (most probably Site I). 



-44- 

Whereas other physical techniques probe extensively, one aspect 
of the catalytic process electron paramagnetic resonance has proven 
successful in probing both the adsorbent and the adsorbate. The 
most serious limitation of epr is that not all catalytic reactions 
involve paramagnetic species; or, even when they do, these species 
might be undergoing strong spin- lattice relaxation that renders 
their detection very difficult. 

References 

1. E. K. Rideal, Concepts in Catalysis , Academic, New York, 1968. 

2. C. R. Adams, Chem. Ind., 1644 (1970). 

3. L. L. Van Reijen, Ber. Bunsen-Ges, 75, 1046 (1971). 

4. J. H. Lunsford, Advan. Catal. , 22, 255 (1972). 

5. J. H. Lunsford, Cat. Rev., 8, 135 (1973). 

6. G. L. Gardner and E. J. Casey, Crt. Rev.. 9, 1 (1974). 

7. J. H. Lunsford, Cat. Rev., 12, 137 (1975); P. H. Kasai and 

K. J. Bishop, Jr., ACS Monograph, No. 171,350 (1976). 

8. L. L. Van Reijen and P. Cossee, Discuss. Faraday Soc., 4l, 

277 (1966). 

9. V. B. Kazanski and J. Turkevich, J. Catal., 8, 231 (1967). 

10. A. Nicula, D. Stamires and J. Turkevich, J. Chem. Phys., 42, 

3684 (1965). 

11. H. Bruins Slot and J. L. Verbeck, J. Catal., 12, 2l6 (1968). 

12. J. H. Lunsford and E. F. Vansant, J. Chem. Soc., Faraday Trans. 

II, 927 (1973). 

13. R. F. Howe and J. H. Lunsford, J. Am. Chem. Soc., 97, 5156" (1975) 

14. A. J. Tench and P. Holroyd, Chem. Commun. , 471 (19"6"8). 

15. Y. Ben Taarit and J. H. Lunsford, J. Phys. Chem., 77, 780 (1973). 

16. K. M. Wang and J. H. Lumsford, J. Phys. Chem., 73, 2069 (1969). 

17. P. Meriaudeau, M. Breysse and B. Claudel, J. Catal., 35, 

184 (197*0. 

18. P. Meriaudeau and J. Vedrine, J. Chem. Soc, Faraday Trans. II, 

472 (1976). 

19. Y. Ono, H. Tokunaga and T. Keii, J. Phys. Chem., 79, 752 (1975). 



-K5- 



Reactions of the Unsaturated Cluster H 2 0s 3 (C0)i O 
With Lewis Bases and Olefins 

erome Keister April 28, 1977 

Metal carbonyl clusters have been the object of increasing 
nterest in recent years due to their potential as unique homogeneous 
atalysts, for novel reactions in which more than one metal center 
lay participate , ' and-, for models for the active sites of metal surface 
.atalyzed reactions. However, although many reactions of clusters 
iave been described, very little is known about their mechanisms due 
o the severe conditions required to induce reactivity with saturated 
:lusters. The reactions of H 2 Os 3 (CO) 10 , a triangular cluster with 
. formal metal-metal double bond, have been studied because its much 
ligher reactivity allows reactions to be carried out under mild 
;onditions and mechanisms can be examined in detail. 

Dihydridodecacarbonyltriosmium reacts with Lewis bases, including 
:arbon monoxide, phosphines, phosphites, arsines, pyridines, iso- 
:yanides, and halides, to form complexes H 2 0s 3 (CO ) i L. 2 These adducts 
;ach have one terminal and one bridging hydride, which undergo intra- 
lolecular exchange with a free energy of activation dependent on the 
tonor properties of L. The mechanism involves a pseudo-four-fold 
•otation of the two hydrides, one axial carbonyl, and one equatorial 
:arbonyl, all on the same metal atom. The adducts H 2 0s 3 (CO) 2 (CNR) 
fe=Me, CH 2 Ph, tBn) undergo intramolecular rearrangement to 
i0s 3 (CNHR) (CO) 10 3 which have both the hydride and C=NHR group bridging 
me edge. 

Reactions of H 2 0s 3 (C0) lo with 1-alkenes form H0s 3 (y-CH=CHR) (CO) 10 
md one equivalent of the corresponding alkane.3 With fumarate esters, 
;he intermediates H0s 3 [CH(C0 2 R)CH 2 C0 2 R] (CO ) x can be isolated and 
characterized. Alkyls can also be isolated from reactions with 
icrylates and moleic anhydride and with ethyl diazoacetate the complex 
I0s 3 (CH 2 C0 2 Et ) (CO) 10 is isolated. These alkyls decompose in the 
>resence of 1-alkenes to form the saturated ester and the corresponding 
I0s 3 (y-CH=CHR) (CO) 10 . In the presence of H 2 the decomposition forms 
I 2 Os 3 (CO) 10 and the saturated ester. Thus, H 2 0s 3 (C0)i O is a catalyst 
"or the hydrogenation and isomerization of 1-alkenes. c 

References 

L. R. B. King, Pro g. Inorg. Chem . , S. J. Lippard ed . , 1_5, 287(1972). 

>. J. R. Shapley, J. B. Keister, M. R. Churchill, and B. G. DeBoer, 
J. Am. Chem . Soc . , 97, 41^5(1975). 

]. a. J. B. Keister and J. R. Shapley, J. Organometal . Chem . , 
85, 029(1975). 

b. J. R. Shapely, S. I, Richter, M. Tachikawa, and J. B. Keister, 
J. Organometal . Chem . , 9*[ 3 C43U975). 

\. J. B. Keister and J. R. Shapley, J. Am. Chem. Soc, 9$, 1056(1976). 



THE REACTIONS OF UNSATURATED TRIOSMIUM 
CARBONYL CLUSTERS 



Mamo'ru Tachikawa 



(Final Thesis) 



May 3, 1977 



1 



Due to the severe conditions necessary to dissociate carbon 
monoxide from the parent molecule, most of the previously known 
triosmium carbonyl complexes derived from the reactions of 
Os 3 ( CO ) i 2 wi th organic molecules were already thermally stable, 
i.e., isolation of conceivable intermediates was very difficult 
Use of the electronically unsaturated cluster H 20s 3 ( CO ) -j q [ I ] as 
the starting material made it possible to prepare otherwise 
unisolable thermally unstable complexes. 2 

From the reactions of [I] with a number of 1,3-dienes, 
compounds of the formula 0s 3 (CO) -j q ( d i ene) were isolated. Depending 
on the positions of substitutions of the dienes, s-cis-l,3-dienes 
on one osmium as well as s-trans-1 ,3-di enes on two osmiums of the 
triosmium unit were produced from this reaction. Of the molecules 
having coordinated s-ci s-1 ,3-dienes , again depending on the posi- 
tions of the substitution, some take axial -equatori al , and others 
take "capped" coordination on one osmium center. 1 ,3-butadiene , 
which is the simplest 1,3-diene, afforded all three of the possible 
forms of coordination, and the two s-cis-diene isomers were found 
to be exchanging on the nmr time scale. 3 



and n 
among 
s e r i e 
R = E 
to fo 
d i p h e 
compo 
which 
di phe 
media 
produ 
of et 
under 
of et 
compo 
appea 
0s 3 (C 



Reaction 
ona- carb 

them wer 
s of osmi 
t. The f 
rm 0s3(C0 
nyl acety 
und that 

could be 
nyl acetyl 
tes betwe 
ct 0s 3 (C0 
hylacetyl 
went ther 
hyl subst 
u n d s with 
red to pr 
0) g (dieth 



s of a 
onyl t 
e 0s 3 ( 
acyc 10 
ormer 
)9(PhC 
1 ene , 
rearra 

b t a i 
ene . 
en the 
MC4P 
ene an 
mal de 
i t u t i 

x i d a 
oceed 
yl eye 



cetyl enes 
r i s m i u m 
CO)io(PhC 
hexadi eno 
was shown 
2 Ph) , whi 
resulting 
n g e into 
ned from 
This demo 
starting 
h 4 M Th 
<T[I], of 
compositi 
n , cycl op 
tively ad 
via unsat 
1 p e n t a d i 



with 
compl e 
2Ph) f 
ne com 

to lo 
ch rea 

i n an 
a know 
the re 
ns trat 

compo 
e seri 

three 
on to 
entadi 
ded me 
urated 
enone ) 



[i] yi 

xes . 
r m d i 
pi exes 
se car 
cted w 

unsta 
n osmi 
action 
ed the 
und Os 
es of 

i some 
give, 
enyl - , 
thyl en 

inter 



el ded 
Of pa 
pheny 
, 0s 3 
bon m 
i th a 
ble 
urn cl 
of 
pres 
3(C0) 
compo 
ri c f 
depen 

and 
e C-H 
media 



a variety 
r t i c u 1 a r i 
1 acetyl ene 
(C0)g(C 4 H 2 
onoxide by 
nother mol 
s 3 (C0)9(Ph 
uster 0s3( 
s 3 (C0) l2 w 
ence of th 
12 and the 
unds from 
orm 0s 3 (C0 
ding on t h 
cycl openta 

bonds. T 
tes of the 



4 Ph 4 



) 



of deca- 
nterest 
, and , a 
R2C0) 's, 

heati ng 
ecule of 
C 2 Ph) 

C0)g( 
i th 

ree inter- 

f i nal 
the reaction 
) g (C4H 2 Et 2 C0), 
e position 
d i e n n e - 
hese reactions 

formul a 



The reaction of [I] with ethylene in cyclooctene resulted in 
a high yield of the labile complex OS3 ( CO ) -j q ( CsH-j 4) 2 [I I ] , which 
could be converted to the slightly more stable 0s 3 (CO) 1 q (NC-CHo^ 
complex[I II] quantitatively. 5 Starting from [II], a large number 
of Os 3 ( CO ) 1 QL2 compounds (L = phosphines, phosphites, nitriles, 
and isoni tri les) , and HOs3(CO)io*X compounds (X = OR, NHR, Cl , 
Br, SR, and H) were prepared in high yield under mild conditions. 



Under somewhat more severe conditions (50-80°C, 10 min.), [HI] 
reacted with a number of nonconjugated dienes, affording 
OS3 (C0)-| o(di ene ) compounds (diene = 1,5-COD, NBD, bi cycl ooctadi ene , 
1 ,b-hexadiene , and, tetraf 1 uorobenzobicycl ooctatri ene) in good 
yield. 6 Under similar conditions, oxidative addition reactions 
of the C-H bonds of various substrates proceeded with the bis- 
acetonitrile complex. Thus, the reactions of pyridine,? acetophenone , 
and, acetaldehyde resulted in H0s 3 (CO ) -, «• X complexes (X = 2-pyridyl, 
2-acetyl phenyl , and acetyl). Another type of reaction of [III] 
was the dehydrogenation of alkyl amines resulting in the formation 
of [I] and HOs^ (CO) , (ami ne-3H) compounds (amine = triethyl amine, 
pyrrolidine, and pi peri dine). 

References : 



1. (a) R.P. Ferrari, G.A. Vaglio, 0. Gambino, M. Valle, and 

G. Cetini, J. Chem. Soc. Dalton, (1972) 1998. 
(b) A.J. Deeming and M. Underhill, J. Chem. Soc. Dalton, 

(1974) 1415. 

2. (a) J.B. Keister and J.R. Shapley, J. Organometal . Chem., 8_5 

(1975) C29. 

(b) A.J. Deeming, S. Hasso and M. Underhill, J. Organometal. 
Chem. , 80 (1974) C53. 

3. M. Tachikawa, J.R. Shapley, R.C. Haltiwanger, and C.G. 
Pierpont, J. Amer. Chem. Soc, 98_, 4651 (1976). 

4. M. Tachikawa, J.R. Shapley, and C.G. Pierpont, J. Amer. 
Chem. Soc. , 97, 7172 (1975) . 

5. M. Tachikawa and J.R. Shapley, J. Organometal. Chem., 
124 (1977) C19. 

6. M. Tachikawa, S.I. Ri enter, and J.R. Shapley, J. 
Organometal. Chem., ]_28 (1977) C9. 

7. C.C. Yin and A.J. Deeming, J. Chem. Soc. Dalton, 
(1975) 2092. 






■ 

TABLE OF CONTENTS 

INORGANIC SEMINAR ABSTRACTS 

1977-1978 

Summer Session Page 

DETERMINATION OF THERMODYNAMICS OF ACID-BASE REACTIONS 1 

USING GAS-LIQUID CHROMATOGRAPHY (glc) - Craig S. Chamberlain 

AMINE-CHELATED LITHIO CYCLOPENTADIENYL TRANSITION METAL 3 

COMPLEXES - Robert Mink 

Fall Ses sion 

THEORETICAL INVESTIGATION OF HOMOGENEOUS ZIEGLER-NATTA 6 

CATALYSTS - David R. Gard 

RECENT DEVELOPMENTS IN THE STUDY OF ISONITRILE INSERTION 9 

REACTIONS - Daniel S. Foose 

ESR AND THERMODYNAMIC STUDIES OF THE REACTION BETWEEN 13 

COBALT (II) PROTOPORPHYRIN IX DIMETHYLESTER AND DIOXYGEN - 
John Breese 

HOMOGENEOUS CATALYSIS OF THE FISCHER-TROP SCH REACTION - 
William L. McCullen 

MECHANISTIC STUDIES OF ALKYL-ALKYL COUPLING MEDIATED BY 18 

ORGANOGOLD COMPOUNDS - James J. Welter 

F-ORBITAL CHEMISTRY - George M. St. George 21 

THEORY AND APPLICATIONS OF EXTENDED X-RAY ABSORPTION FINE 23 
STRUCTURE ANALYSIS (EXAFS) - Paul Young 

MECHANISTIC STUDIES OF THE OLEFIN METATHESIS REACTION - 25 

Bruce Calvert 

EXPERIMENTAL APPLICATIONS OF THE ANGULAR OVERLAP 27 

MODEL - Kenneth Leslie 

CHEMISTRY OF THE EXCITED STATE OF TRIS ( 2 , 2 ' -BIPYRIDINE)- 27 
RUTHENIUM( II) -Nancy P. Forbus 

THE PARTIAL MOSSBAUER PARAMETER AND THE TRANS INFLUENCE 30 

IN LINEAR GOLD(I) COMPOUNDS - Michael J. Desmond 

POLYMER BOUND SCHIFF BASE COMPLEXES - John Gaul 32 



2- 



F all Session (cont'd) 

TEMPERATURE PROGRAMMED DESORPTION AS A TOOL IN 3 3 

CATALYSIS - Steven J. Hardwick 

THE MIXED VALENCE CHEMISTRY OF SOME DISCRETE POLYNUCLEAR 36 
IRON AND RUTHENIUM COMPOUNDS - Susan L. Lambert 

"METAL-QUINONE COMPLEXS" - Michael W. Lynch 3 9 

SYNTHESIS AND STERODYNAMICS OF CYCLOPENTADIENYL 44 

RHODIUM AND IRIDIUM CLUSTER COMPLEXES - 
R. Joe Lawson 

POLYMER SUPPORTED HOMOGENEOUS CATALYSTS - David Corbin 4 5 

THE BINDING OF DIOXYGEN TO METAL CENTERS - 

D. Andre' d' Avignon 49 

SOME ASPECTS OF THE CHEMISTRY OF HYDRIDO BISCYCLOPENTADIENYL 
METAL COMPOUNDS - Greg Pearson 51 

APPLICATIONS OF PHASE TRANSFER CATALYSIS TO ORGANOMETALLIC 
SYSTEMS - Paul C. Adair 52 

ELECTRON DYNAMICS IN BINUCLEAR COPPER(ll) COMPLEXES WITH 

EXTENDED BRIDGING GROUPS - Timothy R. Felthouse 54- 

SYNTHESIS AND STEREODYNAMICS OF TETRAIRIDIUMDODECACARBONYL - 
Gordon F. Stuntz 57 

ELECTRON EXCHANGE ACROSS CYANIDE AND CYANIDE-LIKE BRIDGES 

IN COPPER(ll) SYSTEMS - David S. Bieksza 59 

PART ONE: BINUCLEAR PYRAZINE-BRIDGED COPPER(ll) AND 

VANADYL COMPLEXES PART TWO: CONSEQUENCES OF HYDROGEN 

BONDING INTERACTIONS FOR COBALT-DIOXYGEN COMPLEXES - 

J. Patrick Cannady 60 



Please check the proper box and return to 

." • ' -Ms ..Mary McCabe 

Department of Chemistry 
'• University of Illinois 
..'.'. Urbana, IL 6l801 - : - 



I wish to continue receiving Inorganic Seminar 
abstracts from the Department of Chemistry. 



I do not wish to continue receiving Inorganic Seminar 
abstracts from the Department of Chemistry. 



(Name ) 
(Address ) 



Determination of Thermodynamics of Acid-Base 
Reactions Using Gas-liquid Chromatography (glc) 

Craig S. Chamberlain Final Seminar June 21, 1977 

One approach to understanding the reactivities of transition 
metal complexes is to determine and correlate thermodynamic parameters 
for reactions between metal complexes(acids) and ligands (bases ). * 
However, these parameters are not always easily determined by convent- 
ional techniques when the ligand is gaseous, the metal complex air 
sensitive or available only in small amounts, or when more than one 
ligand tends to coordinate to the metal complex. 2 For this reason, the 
glc method has been investigated as a complimentary alternative to the 
other techniques. 

The basic glc exp res sion for the equilibrium constant for the 
simple reaction A + B* — >A-B is given by 3 




Ki = 

where Kp is the volume of carrier gas required to elute the injected 
ligand from a column containing 1.0 ml of a solution of complex of 
concentration [A] and K°p> the elution volume for a solvent column 
AH and AS are determined by varying the temperature of the experiment, 
i.e., -R In k, = AHi - AS. Contributions to the observed K R values from 
interactions of the ligand at the liquid and solid surfaces must be 
eliminated by one of several methods 4 , yielding fully corrected values. 

The basic theme of the investigations described below is the 
determination of (1) the validity of the various glc methods, (2) the 
limitations of same, and (3) modifications necessary to make the 
methods applicable to systems involving metal complexes. 

A gas chromatograph was constructed and interfaced to an IBM 
Systems 1800 computer. The neat base on column method 5 was tested 
by determination of enthalpies for several hydrogen bonding systems 
using both glc and alternate techniques. 6 Good agreement was found 
between the two sets of data, verifying the validity of this glc method 

To determine if gaseous reactants would present problems the glc 
enthalpy for sulfur dioxide and di-n-octyl ether was determined. There 
is good agreement between this and that calculated using the E and C 
equation. 1 A Schlenk type apparatus has been designed and tested which 
allows one to place air sensitive compounds on the glc column without 
decomposition. 

The solution on column method 3 was then tested since it is the 
method of choice for transition metal complexes. However, it has never 
been coupled with fully corrected Kp values and no comparison of 
enthalpies obtained by this and other methods has been made. Spectro- 
scopic and glc thermodynamic measurements for the interaction between 
bis (7 ,7 ,11,11-tetramethyl heptadecane-8 ,10-dionato ) nickel ( II ) arid 
triethyl amine were made, the enthalpies being in very good agreement . 
Several authors have reported discrepancies between glc and other types 
of equilibrium constants 7 , but in this study the two sets agreed fairly 
well. During this work a simplified graphical technique of obtaining 
fully corrected data was discovered which now makes the glc technique 
available to workers with access to a research grade chromatograph. 



Several transition metal complexes have been prepared for this work 
which have the unusual property of being high molecular weight 
liquids at room temperature. 

Difficulties are encountered using conventional techniques to 
determine thermodynamic parameters for the first step in multi- 
step system^ of the type 



B 



B 



A 



^ AB 



-* AB; 



Ki,AH x 



K 2 ,AH 2 



because of contributions from the second step. 2 Bis (di-n-octyl- 
dithiophosphato) nickel(II) will coordinate two moles of pyridine. 
Basic glc equations indicate that only contributions for the first 
interaction should be seen on the glc if small injection volumes of 
base are used. The glc enthalpy was obtained using this method. It 
compared well to the AH X reported for essentially the same system 
obtained by a spectrophotometric method. 8 Thus it appears that glc 
can be used to obtain &H X for systems of this type. 

The above studies indicate that glc is a reliable technique which 
is applicable to difficult systems. Based on the characteristics of 
the glc technique a plan is proposed to determine the enthalpies for the 
second step of the following interaction for a number of non-volatile 
bases B 1 : 



B 



B 



A 



^AB 



=^ABB T 



Kx.AKj 



K 2 ,AH : 



"ffhis information would allow one to learn more about how B' effects 
the acidity of AB' towards B. This is, of course, important in 
certain catalytic and biological systems. 



References 

1. R. S. Drago, Structure and Bonding, 15 73 (1973). 

2. T. 0. Maier and R. S. Drago, Inorg. Chem. , 11 1861 (1972). 

3. D. F. Cadogen and J. H. Purnell, J. Chem. Soc . A 1968 2133- 
l\. H. L. Liao and D. E. Martire, Anal. Chem. ^4 ^98 (1972). 

5. H. L. Liao and D. E. Martire, J. Amer. Chern. Soc. 9§_, 2058 (197*0 

6. C. S. Chamberlain and R. S. Drago, J. Amer. Chem. Soc., 98 , 
6l*J2 (1976). 

7. D. E. Martire, Anal. Chem. .46, 1712 (197*0. 

8. R. L. Carlin, et al., Proc. Chem. Soc. 196*4, 228. 



Amine-Chelated Lithio Cyclopentadienyl 
Transition Metal Complexes 

Robert Mink (Final Seminar) August 19, 1977 

Organolithium chemistry has experienced a rapid growth in 
recent years due to the large synthetic utility shown by the 
reagents. In many situations, organolithium reagents are 
becoming preferred to Grignard reagents. Studies have shown that 
the reactivity of an organolithium compound increases greatly 
when the lithium atom is complexed with tertiary polyamines, 
the most common being N, N, N', N' -tetramethylethylenediamine 
(TMED) . 

The metalation of ferrocene using n-butyllithium yields a 
mixture of the mono- and dilithio complexes. Pure monolithio- 
ferrocene can only be prepared from a substituted ferrocene. 
In the presence of a chelating agent such as TMED or 1, 1, 4, 
7, 7-pentamethyldiethylenetriamine (PMDT), the metalation of 
ferrocene produces pure amine-chelated dilithioferrocene. The 
X-ray structure and NMR spectra of dilithioferrocene- PMDT 
confirm the presence of only one PMDT moiety per two lithium 
atoms. NMR data indicates that this compound exhibits stereo- 
chemical nonrigidity with the structure in solution identical 
to the solid state structure at low temperatures. From the 
NMR spectra it was determined that two exchange processes are 
occurring; one involving the PMDT and cyclopentadienyl rings 
and the second involving the PMDT only. The mechanism suggests 
an attack of PMDT from the solvated to the unsolvated lithium 
atom. At higher temperatures the PMDT completely dissociates. 

It was found that the base PMDT is more advantageous 
than TMED because of the greater solubility in aromatic 
solvents of the resulting amine-chelated lithio transition 
metal complexes. The metalation of (C 5 H 5 ) 2 ReH using 
n-butyllithium and PMDT produces ( C 5 H 5 ) 2 ReLi • PMDT, where 
lithiation has occurred at the metal atom rather than the 
cyclopentadienyl ring as in ferrocene. The rhenium-lithium 
complex reacts with carbon monoxide, carbon dioxide, water, and 
hydrogen. The treatment of (C 5 H 5 ) 2 ReLi- PMDT with alkyl halides 
leads to the formation of (C 5 H 5 ) 2 ReR (R = alkyl) complexes, 
which show high thermal stability. The reaction of 
(C 5 H 5 ) 2 ReLi- PMDT with organometallic halides yields dimetallic 
complexes in which a metal-metal bond is formed. Two of the 
dimetallic compounds also contain a bridging carbonyl group. 

Addition of n-butyllithium and PMDT to bis (cyclopentadienyl )- 
molybdenum dihydride affords bis ( cyclopentadienyl )hydrido 
(1, 1, 4, 7, 7-pentamethyldiethylenetriaminelithio)molybdenum. 
The PMDT- lithio tungsten analog has also been prepared but, like 
the TMED- lithio cyclopentadienyl transition metal complexes, 
it is insoluble in aromatic solvents. Both the molybdenum- 
lithium and tungsten-lithium intermediates react with carbon 
monoxide, carbon dioxide, hydrogen, and alkyl halides. 



Theoretical Investigation of 
Homogeneous Z iegler-Na t ta Catalysis 



David R. Gard 



October 11, 1977 



Int roduc t ion 

< 1 
Ziegler's report in 1954 that ethylene could be polymerized 

at atmospheric pressure and ambient temperature to a linear, high 

molecular weight polymer with a combination transition metal-metal 

alkyl catalyst opened an entirely new field of catalysis. Other 

workers have considerably extended this reaction, in particular 

Natta, whose control of the s ter eor egular polymerization of 

a-olefins with Zielger catalyst systems has led to wide industrial 

application . 



Th 
mechani 
complex 
benefit 
tions h 
propose 
and has 
alkylat 
coordin 
alkyl g 
transit 



ere have bee 
sm of Zielge 
ity o f these 
. ' Nevert 
ave revealed 
d by Cossee 

received wi 
ion of the t 
ation to an 
roup to the 
ion state: 



n numero 
r-Natta 

systems 
heless , 
the mos 
concurs 
de accep 
r ansi t io 
oc tahedr 
olefin t 



us suggestions proposed as to the 
catalysis, but due to the diversity and 

many of these ideas are of limited 
a great number of experimental investiga- 
t fundamental features. The mechanism 

most fully with the experimental data 
tance. ' This mechanism involves 
n metal center followed by olefin 
al vacant site and cis-migr at ion of the 
hrough a concerted, four- cent ered 



X — M— + 

X X 



C 2 H 



R 

I 

X — M 

/I 

X X 



/* 



t 

CH 2 



X 



R 

— M' 

/I 
X X 



-X 



"CH 



CH 



1/ 



X 



-> X — M — CH 2 CH 2 R 
/ I 
X X 



The Cossee mechanism is essentially qualitative, but recently some 
theoretical studies have been performed in order to examine 
quantitatively the electronic structure of the proposed intermediates. 



Chain Propagation via Insertion 

7 9 

Armstrong and Novaro have conducted SCF-CNDO-MO, all valence 

electron calculations of the electronic structure of model soluble 

cataly st-olef in complexes, (I) and (II): 

Me 



CI- 
CI" 



Al 



CI 
CI- 



C1 M 

f Me 
Ti^ CH 2 

c'i /; 

CH 2 



II 



Et 
Et 



Al" 



0. 

•0 ' 
Me 



Bu 

I 
Ti" 

I 



Me 



CH 2 
-\ 
OMe 

Al 

I ^ 
Et 



CH 



Et 



The changes in charge distribution and molecular orbital energies and 
character over the course of the reaction support Cossee's overall 
conception of insertion as the mechanism of chain propagation in 
Ziegler-Natta catalysis. However, certain points of disagreement 
arise which afford a better understanding of the process. The 
bond order of the t i tanium-olef in TT-backdonat ion component = "0.01 
for (I). This casts doubt upon Cossee's proposal that stabilization 
of the titanium f t 2 ' orbital is a prerequisite for the activation 
of the t i taniu'm-alky 1 bond. Hence, the driving force is not prior 
activation of the t i tanium-alky 1 bond, but a substantial lowering 
of the energy barrier to alkyl insertion through proficient mixing 
of the alkyl a and olefin tt* orbitals via the 't 2 ' orbital. The 
t itanium-alky 1 bond is quite polarized, with a charge on the titanium 
atom of +1.17 for (I) and on the alkyl a-carbon atom of -0.24 and 
~0.30 for (I) and (II) respectively. The olefin has an overall 
slight positive charge and the reaction is initiated by the inter- 
action between the olefin and the alkyl group in the catalys t-olef in 
complex. Only minor changes in the orbital energies occur during the 
insertion reaction, with the substrates remaining bonded to the 
titanium atom at all times. The cataly tically active complex is 
trigonal bipyramidal before olefin coordination, and not octahedral 
with a vacant site. This has..been confirmed by EPR * and by 
further calculations on (II). 

Unsymme trically substituted ethylenes with weakly polar double 
bonds such as propene or styrene exhibit predominately Markovnikoff 
insertion with t itanium-alky 1 bonds. With Group VIII metal based 
catalysts, however, a pronounced tendency for ant i-Markovnikof f 
insertion is observed. This r egioselec t ivi ty can be explained 
using the orbital overlap model of Armstrong, which considers the 
relative shifts in population of the olefin tt and tt* orbitals 
upon coordination to the catalyst complex. 

Catalyst Tailoring and (3-Hydrogen Transfer 

Tailoring of Ziegler-Natta catalysts to develop polymers of 
desired molecular weight, degree of branching, or possession of 
unsaturated grou >s depends mainly upon control of 3-hydrogen transfer 
at the transition metal center, as this reaction can bring-about 
termination or isomer izat ion of the growing polymer chain. Novaro 
has attempted to show that 3-hydrogen transfer between a«iiJa«MB*ar ole^iv* 
and alkyl group and insertion stem from the same initial situation 
and that interaction of an alkyl group and an olefin on a titanium 
center can lead to either reaction. 

Cone lus ion 



Molecular orbital calculations have complemented the diverse 
experimental data in providing insight into the nature of Ziegler- 
Natta catalysis. The Cossee mechanism has gained feasibility but 
more work of this type needs to be done to differentiate between 
alternative intermediates and mechanisms both for Ziegler-Natta 
and for other catalytic systems. 



References 



1. Karl Ziegler, E. Holzkamp, H. Breil , and H. Martin, Angew. Chem. 
67, 426, 541 (1955). 

2. G. Natta and I. Pasquon, Adv. Cat. 11, 1 (1959). 

3. J. Boor, Jr., Macromol. Rev. _2, 115 (1967). 

4. Max Herberhold, "Metal TT-Complexes , " Elsevier, Amsterdam, 1974, 
Vol. II, part 2, p. 244. 

5. W. Cooper, "Comprehensive Chemical Kinetics," C. H. Bamford and 
C. F. H. Tipper, eds., Elsevier, Amsterdam, 1976, Vol. 15, 
Chap . 3 , p . 13 3. 

6. P. Cossee, J. Cat. _3> 80 (1964); Rec. Trav. Chim. 8_5, 1151 
(1966); "The Stereochemistry of Macromolecules , " A. D. Ketley, 
ed., Marcel Dekker, New York, 1967, Vol. I, chap. 3, p. 145. 

7. D. R. Armstrong, P. G. Perkins, and J. J. P. Stewart, 
J. Chem. Soc. Dalton 1972 , 1972 (1972). 

8. A. K. Ingerberman, I. J. Levine, and R. J. Turbett, J. Polymer 
Sci. Al _4, 2781 (1966); J. W. Begley and F. Penella, J. Cat. 8, 
203 (1967). 

9. 0. Novara, S. Chow, and P. Magnouat , J. Cat. 41 , 91 (1976). 

LO. T. S. Djabiev, R. D. Sabirova, and A. D. Shilov, Kinet. Katal. _5, 
441 (1976). 

0. Novaro, S. Chow, and P. Magnouat, J. Cat. 4J2, 131 (1976). 

G. Henrici-Olive and S. Olive, Top. Curr. Chem. 6J7_, 107 (1976). 

G. Henrici-Olive and S. Olive, Angew. Chem. Int. Ed. H), 105 
(1971), and references therein. 



RECENT DEVELOPMENTS IN THE STUDY OF ISONITRILE 
INSERTION REACTIONS 



Daniel S, Foose 



October l8, 1977 



The insertion reaction is one of the most important reactions in 
organometallic chemistry. This type of reaction is one that conforms 
to the general equation 



M-X + Y 



M-Y-X 



where M is a metal and X and Y are monatomic or polyatomic species. 
The recent growth of interest in these reactions stems from the reali- 
zation that homogeneous catalytic reactions, such as hydroformylation 
or polymerization, proceed by one or more insertion steps. In addition, 
many insertion reactions have been successfully employed in organic and 
organometallic synthesis. The insertion of isonitriles into metal- 
carbon bonds was first reported by Yamamoto, et al. 1 in 1968. The large 
amount of work that appeared in the years immediately thereafter has 
been reviewed. 2 > 3 -> 4 Recent investigations have been concerned with 
isonitrile insertions into metal-carbon, metal-hydrogen, metal-halogen 
and metal-metal bonds. 

Isonitriles are known to insert into metal carbon bonds involving 
nickel, palladium, platinum, iron, molybdenum and manganese. The first 
isonitrile insertion products obtained 1 > 5 were TT_cyclopentadienyl 
(cyclohexylisonitrile) [alkyl (cyclohexylimino) methyl] nickel. 

n-C 5 H 5 Ni(PPh 3 )R + 2R'NC -> n-C 5 H 5 Ni(CNR' ) [C(=NR r )R] + PPh 3 

No intermediate was isolated and the mechanism of the reaction was not 
ascertained. Several other examples of both single and multiple iso- 
nitrile insertions have since been reported. The complex trans- 
iodobis(dimethylphenylphosphine) methyl palladium is known to undergo 
both single and multiple insertions into the palladium-carbon bond® to 
produce products of types I, II, and III. Recent studies 7 of related 



1 PPMCH,), 

\ / 

Pd 
/ \ 

(CH^PhP CCH, 

NC 4 H„ 



1 PPhCCH,), 
\ / 
Pd 



■'1 



|C|— CH, 
NC.H,, 



II 



NC.H,, 

lCH,)Ph 3 P C-C=NC.H ll 

Pd I 
/ \ I 
I N=CCH, 

C.H n 

III 



compounds have shown that the steric effect of the phosphine, iso- 
nitrile, or alkyl ligands seems to be an important factor in the control 
of reactivity, for both multiple insertions as well as single insertions. 
Other investigations of palladium-carbon insertions include the reactions 
of dimethyl- and diphenyl-mercury with dihalobis( isonitrile) palladium 
complexes to produce bis(imino) compounds 8 and a study of the mechanism 
and stereochemistry of insertion into metal alkyl and vinyl bonds in 
square planar complexes. 9 Multiple insertion products have been ob- 
tained with the direct reaction of isonitriles with dicarbonyl-TT- 
cyclopentadienylalkyliron. Structure IV was proposed as a possible 



structure for the triple insertion product "> , ra . „ 

^ ™ Cl " A revised structure 11 




o, 



CO 

/ 

e. 



'CH 3 J 3 CNH-C I 6H " 

PhCH,C- -6=NR \ c ^_, MC(>Hit 

iv ,» 

for this "trisimino-type comm py" hp c v. 

structural determination naTrevealed I ' '? 7 + been sported. X-ray 

carbene (structure v) . revealed a novel type of chelating 

has beeTo^ne^om^hf Actions o^anio^ ,"»*"— and manganese 
plexes of the general type I (rm) m~ °f. a " lonic °rganometallic com- 
bat contain a^ovel'dinlp\MSo4li1gand^ni v r b 12 C °T 6XeS 

J j-j-feciiiu ^vi; nave been produced. 

N ^ N R 



R 



VI 



R 

VII 



iganas^o^orm X^CbS^ff^*^ -act with donor 

igands (VII). 19 contain normal monohapto iminoacyl 

VI + L - VII • L 



j^in^ as if they are 

ransformation. An interesting feature If thf 1 ? dlha P to -^riohapto 
hat it contains a lone pair on L t! he Jl'-^^oacyl VTl) is 

3 electrophilic attack nitrogen atom that is susceptible 

R' 

/J~ R )»— * 

M — C + Rf X -> M<— (T 

R \ R 

VIII 

{lecules i0 cont r a?ninl a the n a ^ thls h site ^ to the formation of 

two ison?trxle ligand-- in n ?h"^^ llg , and (VIIl) " The Presence 
; om permits the occurrence of 1 " 1Mtl0n s P here of the metal 
a-rangements lS x rav w, ""Itiplo isonitrile insertion re- 
X-ray structural analysis of (i X ) shows the presence 

^-CsH 5 ) t ,o(CO)(c N CH 3 ) 2 - + 2CH 3 I > (r,= -C s H 5 )Mo(co) (CNCH 3 ) 2 (CH 3 ) 2 I 

IX 



11 



of a polyhapto ligand that can Hp Hp,,^^ 

aminocarbene. described as an iminodimethyl- 

Insertions of isoni t-r-i i o«- -s *»* 
reported for complexes S pUtii5E°i«?« X "J!5" lrl(ta bondE ^ve been 
rPtH(CNR)(PEt 3 )]Cl undergoes ins^iioai nto h ^ C ° m ? leX trans " 
to afford complexes co^^^^aS^i^^gSJScSg 

I / 'I / 

■ Pt-i C Pt C 

I V_ I \ 

/ 

JSjlSS £ -"3^*^^^^ W«. or two isomers, 
secondary carbene products a «ords relatively uncommon cationic 

bond/or"^ 2S\££E SSSL^JSj^^ ^ -tal halogen 
have recently been extended to include the h^T TheSe ins ertions 
ium, zirconium, hafnium and vanadium"?^ 6 com P lex es.af titan- 

Recent examples involving the inqprti^ ~-r • -^ . 
metal bonds are the reactions of Lo^huc * isonitriles into metal- 

[(c 6 H 5 ) 2 ] 2 ][ P p 6]a and [pt cnot! K 1 firSN^i' !^)^ 

spicuous feature of this reaction U ?h= 2 2][ ? s]2 - The most 0o «- 
distance which occurs upon the formation P H S1 °\ 0f the meta l »etal 
complexes. y lormation of these bridging isonitrile 

Conclusion: 

The increasing number of renort* of !■,„„«*■•, 
metal-carbon, metal-hydro F en mlt? L?f" omtrile insertions into 
to indicate the generality of £his reaction tn meta l-metal bonds seems 
Recent investigations have shown the In o tanc, nTT etaUiC chemistry, 
both single and multiple insertion" IS "5 + S Steric control in 
nitrile insertion products It the Mhtl+ -^ react ^ity of iso- 
may provide us with an increlsfa'underst^dlnc, 1 ^ v£° yl c °^™*, 
of isonitrile insertion reactions g the overal l course 



References : 
1, 



li, ^3^(1968?: YamaZaki and N ' Hagihara, Bull. Chem. Soc. Jap., 

2. Y. Yamamoto and H. Yamazaki, Coord. Chem. Rev., 8, 225 (1972). 

3. P. M. Treichel, Adv. Organometal. Chem., 11, 21 (1973). 

**. F. Bonati and G. Minghetti, Inorg. Chim. Acta, 9, 95 (1974). 

5 * Wtl969h' H * YamaZaki ' N « Ha S ih ^a, J. Organometal. Chem., 18, 

6. Y. Yamamoto and H. Yamazaki, Bull. Chem. Soc. Jap., 43, 2 6 5 3 (l 97 o) 



12 



7. Y. Yamamoto and H. Yamazaki, Inorg. Chem. , 13, 438 (1974). 

8 ' ?l976) Ciani and M * Nicolini ' J - Organometal. Chem., 104, 259 

9. S. Otsuka and K. Ataka, J.C.S. Dalton, 327 (1976). 

10. Y. Yamamoto and H. Yamazaki, Inorg. Chem., 11, 211 (1972). 

11. Y. Yamamoto and K. Aoki, Inorg. Chem., 15, ^8 (1976). 

12 ' cil D (l976? S and D ' F ' Chodosh ' J * Organometal. Chem., 122, 

13. R. D. Adams and D. F. Chodosh, J. Amer. Chem. Soc., 98, 5391 (1976) 

14 - L^^^d 1 ^)^' ^ "• *' StePanlak ' J - <*■»— *• 

15 ' C9 U975)" iStian and H ' C ' Clarkj J- Or e anometa l- Chen., 85, 

L6. B. Crociani and R. Richards, J. Chem. Soc. Chem. Comm. , 127 (1973). 

11 ■ ?6l? r ci ia i975^' NiCOlini and R - Rlcha rds, J. Organometal. Chem., 

99,°5502 ea i97?) H ° PSj L ' Benner ' A ' Balch > J - Amer. Chem. Soc., 

19. R. D. Adams and D. F. Chodash, J. Amer. Chem. Soc., 99, 6$kh (1977) 



L8 



13 



ESR AND THERMODYNAMIC STUDIES OF THE REACTION BETWEEN COBALT ( II ) 
PROTOPORPHYRIN IX DIMETHYLESTER AND DIOXYGEN 

John Breese (Final Seminar) October 20, 1977 

The oxidation of organic substrates by molecular oxygen is an 
extremely important class of chemical transformations. The conversion 
of petroleum-based chemicals to useful chemical intermediates by con- 
trolled oxidation is a ma.jor concern of the chemical industry. Like- 
wise, the inhibition of the oxidative decomposition of numerous 
"finished products" such as plastics, lubricating oils, pai'nts, and 
rubber is of equal importance. Catalysis of the first process and 
inhibition of the latter by transition metal complexes have both been 
areas of considerable technological interest. 1 " 5 

In this study 'a systematic determination of the factors important 
in the binding and activation of dioxygen by a transition metal complex 
was undertaken. Also the electronic perturbation that the binding of 
2 has on the metal center was considered. Cobalt (II ) complexes are 
particularly well qualified for understanding the electronic nature 
of the metal-0 2 interaction as they have at least one unpaired elec- 
tron, making them ESR active, and, in many geometries, an unpaired 
electron resides in the orbital to which the dioxygen binds. The ESR 
analysis of the cobalt-dioxygen complexes has been the source of some 
speculation and debate in the literature. G ~ 10 Initially, it was hoped 
that this study would shed some light on many of the present 
misconceptions . 

Cobalt (II) protoporphyrin IX dimethylester (CoPIXDME) was the 
complex chosen for this study. The enthalpies for dioxygen binding 
to this complex as a function of base have previously been reported. L1 * : 
The enthalpy for the reaction 

CoPIXDME + B - CoPIXDME- B 

was measured for the bases 1-methyiimidazole, pyridine, piperidine, 
and tetrahydrothiophene. A good fit to the E and C equation was ob- 
tained, which implies that TT-backbonding is unimportant in these com- 
plexes. It was found that there is a direct relationship between the 
strength of base binding and the strength of oxygen binding. 

The fractional electron transferred (E. T. ) from the cobalt to 
the oxygen was measured as a function of axial base by a previously 
reported procedure. 13 It was found that as the base-cobalt bond gets 
stronger the E.T. decreases, which is exactly the opposite behavior 
from that predicted by the "electron transfer model." 14 This obser- 
vation is explained by contributions from three mechanisms: 

1) core polarization 

2) indirect polarization arising from unpaired spin 

residing on the dioxygen fragment 

3) mixing of the doublet ground state with the quarter 

excited state. 15 

By consideration of other cobalt-dioxygen systems, it is shown that 
the 1 latter mechanism is the most imoortant. 



14 



Oxidation studies utilizing 2, 6-dimethylphenol indicate that 
as the base gets stronger (and supposedly the complex has more 
cobalt(lll)-0 2 character) the oxygen becomes more activated. How- 
ever, a limit has been found for the base strength, after which 
irreversible oxidation of the metal is the dominant reaction. 



1 



o 

7 
8 

9 

10 
11 

12 

13 

14 

15 



References 

T. Dumas and W. Bulani, Oxidation of Petrochemicals: Chemistry 
and Technology , J. Wiley and Sons, New York, 1974. 

M. M. T. Khan and A. E. Martell, Homogeneous Catalysis by Metal 
Complexes , Vol. I, Academic Press, New York, 1974. 

R. A. Sheldon and J. K. Kochi, Oxid. and Comb. Revs., 5, 135 (1973) 

M. M. T. Khan and A. E. Martell, Homogeneous Catalysis by Metal 
Complexes , Vol. II, Academic Press, New York, 1974 . 

J. E. Lyons, Advan. Chem. Ser. , 132 , 64 (1974). 

A. L. Crumbliss and F. Basolo, Science, 164 , ll68 (1969). 

A. L. Crumbliss and F. Basolo, J. Amer. Chem. Soc., 92, 55 (1970). 

B. M. Hoffman. D. L. Diemente, and F. Basolo, J. Amer. Chem. Soc. , 
92, 61 (1970). 

B. M. Hoffman and D. H. Petering, Proc. Natl. Acad. Sci. U.S.A., 
67, 637 (1970). 

F. A. Walker, J. Amer. Chem. Soc., 92, 4235 (1970). 

T. J. Beugelsdijk, Ph.D. Thesis, University of Illinois, 1975. 

T. J. Beugelsdijk and R. S. Drago, J. Amer. Chem. Soc, 97, 6466 
(1975). 

B. S. Tovrog, D. J. Kitko, and R. S. Drago, J. Amer. Chem. Soc., 
98, 5144 (1976). 

B. S. Tovrog, Ph.D. Thesis, University of Illinois, 1975. 

B. R. McGarvey, Can. J. Chem., 53, 24 98 (1975). 



15 

HOMOGENEOUS CATALYSIS OF THE FISCHER-TROPSCH REACTION 
William L. McCullen November 1, 1977 

For several years, the Fischer-Tropsch reaction 1 (eq. l) has 

M,MO 

CO + H 2 > alkanes, alkenes, alcohols, and (l) 

other oxygenated products 

M, MO = transition metals and metal oxides 

been employed in South Africa for the industrial production of a wide 
variety of useful organic products from synthesis gas, a mixture of 
CO and H 2 which is readily available from coal gasification. 2 * 3 In 
this respect, coal could supplant crude petroleum as the source of raw 
material for production of industrial chemicals. However, industrial- 
scale implementation of the heterogeneous catalysis of the Fischer- 
Tropsch reaction suffers two major drawbacks. One undesirable feature 
is the lack of reaction specificity. In view of the variety of pro- 
ducts obtained from the reaction shown in eq. 1, it is not surprising 
that the product distribution is quite sensitive to the CO/H 2 ratio, 
reaction conditions, and the selection of the metal catalyst. Despite 
stringent control of reaction parameters (temperature and pressure) and 
judicious choice of the catalyst, the synthesis generally yields several 
products. Difficulties in studying surface properties, chemisorption, 
and adsorbate properties have hampered attempts at determining the ki- 
netics and mechanism of the heterogeneous process. 3 Another significant 
disadvantage is the large energy requirement. Ordinarily, tremendous 
pressure (10-1000 atm. ) and high temperature (150-500°C) are required 
to activate the metal catalyst. Such great energy expenses are be- 
coming increasingly uneconomical with respect to the rising cost and 
availability of energy sources. 

There is reasonable belief 4 that homogeneous catalysis will pro- 
vide greater product selectivity than heterogeneous methods with smaller 
energy requirements. However, metal surfaces exhibit certain properties 
with respect to an adsorbed species which should be preserved in the 
design of homogeneous catalysts. 5 The possibility of multi-metal bond- 
ing interactions between the adsorbed species and the metal surface, 
and the apparent mobility of the adsorbate over the metal surface are 
two such properties which are probably crucial in the activation and 
subsequent reduction of carbon monoxide. Metal clusters have been pro- 
posed as a general class of compounds that might serve as effective 
homogeneous catalysts for the Fischer-Tropsch reaction. 3 " 9 

In designing systems for catalytic behavior, Muetterties has sug- 
gested the hypothesis that the susceptibility of CO to reduction should 
directly correlate with the degree of CO bond lengthening in the metal 
complex. 9 As a consequence, increasing the number of metal-carbonyl 
interactions should lengthen the CO bond. Crystal structures of metal 
carbonyls show this to be generally true. However, the synthesis of 
hydrocarbons probably requires greater bond reduction than can be re- 
alized through any of the three normal modes of CO bonding for it in- 
volves the scission of the CO bond. (This is the strongest bond of any 
diatomic molecule.) To facilitate CO bond cleavage, the following 



16 



bonding modes have been proposed: 9 






Crystallography has never revealed any of these bonding modes for metal 
carbonyl compounds, yet it is likely that such interactions may be 
involved in the transition state species of CO reduction processes. 
The Hy-ti 2 (x=1,2) modes have been observed for various acetylenes in 
Ni 4 clusters and, in a few instances, reaction with H 2 resulted in the 
facile reduction of the coordinated acetylene to cis -olefins. Another 
important feature which suggests the selection of metal clusters as 
homogeneous catalysis is the well-established phenomenon of intramolecu- 
lar ligand mobility. The oxidative addition of H 2 followed by hydride 
migration is probably an important sequence in CO hydrogenation; 
furthermore, CO migration is considered an essential step in the syn- 
thesis of compounds containing carbon-carbon bonds. 

Survey of Homogeneous Fischer-Tropsch Systems 

There have been several reports of homogeneous Fischer-Tropsch 
type reactions, in which the promoter or active catalytic species were 
proposed to be mono-metallic. Bercaw and co-workers 13 reported the 
synthesis of methanol in a three step process promoted by a derivative 
of [ (r) 5 -C 5 Me 5 )ZrN 2 ] 2 N 2 generated in situ . Shoer and Schwartz 14 observed 
the synthesis of a series of low molecular weight alcohols from CO and 
(i-C 4 H9) 2 AlH in a multi-step reaction promoted by Cp 2 ZrCl 2 . Caulton 
and others 15 reported that Cp 2 TiCl 2 promoted the methanation reaction 
under mild conditions. However, the workers did suggest the possi- 
bility of a bimetallic species in the reaction mechanism. Petit 16 
reported the synthesis of methanol catalyzed by Fe(CO) 5 and Cr(CO) 6 . 
Unfortunately, few details are readily available for this reaction. 

Undoubtedly, a great deal of research using metal clusters is 
being conducted but few results have appeared in the literature. Per- 
haps the most significant contribution to date is that of Pruett and 
co-workers in which they report the synthesis of ethylene glycol from 
CO and H 2 promoted by Rn 4 (CO) 12 . 1T Infrared work suggests that the 
clusters [Rh l2 (CO) 34 2 ~ ] and [H 5 _ n Rh 13 (CO) 24 " ] are the dominant re- 
action species. 18 Muetterties and co-workers 19 * 20 have reported homo- 
geneous catalysis of the Fischer-Tropsch reaction with Os 3 (CO) 12 and 
Ir 4 (C0) 12 yielding low molecular weight alkanes under mild reaction con- 
ditions. The use of a molten salt medium and the addition of a particu- 
lar Lewis acid has proven to be a more selective and efficient synthetic 
method than the use of organic solvent media. Though the reaction 
mechanisms for these systems are unknown, it is believed that inter- 
action of the CO moiety with several metal centers in an asymmetric mode 
is responsible for the activation of the carbon-oxygen bond toward 
hydrogenation. 



17 

Summary 

Homogeneous catalysis of the Fischer-Tropsch reaction should 
provide a more efficient and selective synthetic means of chemical 
production than heterogeneous catalysis, the method primarily used 
by industry today. The design of homogeneous catalysts should com- 
bine features of metal surfaces with the versatility of organo- 
metallic compounds. Metal clusters should provide such a class of 
compounds , and preliminary results indicate that some heavy metal 
clusters do exhibit catalytic behavior in the Fischer-Tropsch re- 
action. These findings are encouraging and have stimulated further 
research in this area. 

References 

1. H. Storch, N. Golumbr, and R. Anderson, "The Fischer-Tropsch 
and Related Synthesis," Wiley, New York, N.Y., 1951. 

2. I. Wender, Catal. Rev.-Sci. Eng . , 14 , 97 (1976). 

3. M. A Vannice, Cat. Rev.-Sci. Eng ., 14, 97 (1976). 

4. E. L. Muetterties, Science , 19b , 859*Tl977) . 

5. E. L. Muetterties, Bull. Soc. Chim. Belg , 84, 959 (1975). 

6. A. L. Robinson, Science , 1Q4 , 1150 (197b). 

7. J. R. Shapley, St rem Chemiker , to be published. 

8. A. K. Smith and J. M. Basset, J. Mol. Cat. , 2, 229 (1977). 

9. E. L. Muetterties, Bull. Soc. Chim. Belg ., 85, 451 (1976). 

10. E. L. Muetterties, et. al. , J. Am. Chem. Soc . , 98 , 8289 (1976). 

11. M. G. Thomas, E. L. Muetterties, R. 0. Day and V. W Day, 
J. Am. Chem. Soc, 98, 4645 (1976). 

12. E. L. Muetterties, et. al. , J. Am. Chem. Soc , 99, 743 (1977). 

13. J. M. Manriquez, D. R. McAlister, R. D. Sanner and J. E. Bercaw, 
J. Am. Chem. Soc , 98, 6733 (1976). 

14. L. I. Shoer and J. Schwartz, J. Am. Chem. Soc , 99 , 5831 (1977). 

15. J. C. Huffman, J. G. Stone, W. C. Krusell and K. G. Caulton, 
J. Am. Ch em. Soc , 99, 5829 (1977). 

16. R. Petit, "Homogeneous Catalyst System for the Conversion of 
CO and H 2 to Methanol and Hydrogen," 172 ACS Meeting, 

San Francisco, I NOR 82. 

17. U. S. Pats. 3,833634 (Chem. Abstr . , 79, 78081) and 3,878,214, 
3,878,290, and 3,878, 292" TChem . Abstr . , 83, 45426-45428). 

18. S. Martinengo, B. T. Heaton, R. J. Go'odfellow and P. Chini, 
J. Chem. Soc , Chem. Comm . , 39 (1977). 

19. M. G. Thomas, B. F. Beier and E. L. Muetterties, J. Am. Chem. Soc . , 
98, 1296 (1976). 

20. G. C. Demitras and E. L. Muetterties, J. Am. Chem. Soc , 99 , 
2796 (1977). 



MECHANISTIC STUDIES OF ALKYL-ALKYL COUPLING 
MEDIATED BY ORGANOGOLD COMPOUNDS 

James J. Welter November 3, 1977 




(CH 3 ) 3 AuPPh 

CH 3 CH 3 + Au(metal)+ PPh 3 

The decomposition via coupling of the methyl groups sparked the 
present day interest in mechanistic s tudies of trialkyl gold 
complexes . 

Alkyl groups on organogold complexes have been shown to couple 
with alkyl halides as shown below for CH 3 AuL and CH 3 I. 7 

CHaAu 1 ! + CH 3 I > (CH 3 ) 2 Au III IL (2) 

(CH 3 ) 2 Au II][ IL + CHaAu 1 ], aS > (CH 3 ) 3 Au Ii:L L + IAu 1 ! (3) 
(CH 3 ) 3 Au I:[I L > CH 3 CH 3 + CH 3 Au x L etc. (4) 

NMR studies have helped to elucidate the mechanisms of the 
oxidative addition, alkyl isomerizat ion , and reductive elimination 
steps in this cycle. 

Oxidative Addition 

The rate of the oxidative addition of CH 3 I to CH 3 AuL depends 
on the nature of L. The rate of addition of CH 3 I to CH 3 AuP(CH 3 ) 3 
is approximately five times faster than the corresponding rate for 
CH 3 AuPPh 3 . 8 Reaction (2) at high concentrations of CH 3 I follows 
psuedo first order kinetics. 9 The observed rate, constants for 
eq . (2) were found to be approximately proportional to the CH 3 I 
concentration. A free radical chain process was ruled out by 
the failure of the free radical scavenger galvinoxyl to affect 
the rate of the reaction. 

The synthesis of dialky laurate ( I ) 1 ° species (eq. 5) provides 

RAuPR 3 + R'Li > RR'AuPR 3 Li (5) 

a very reactive anionic organome tallat e which can oxidatively add 
alkyl halide instantaneously upon mixing to give trialkylgold 

species . 

RR'AuPR 3 Li + R"X > RR , R''AuPR 3 (6) 



Alkyl Exchange 

Investigation of the mechanism of the alkylation of dialkyl- 
gold(III) compounds with alkylgold(I) complexes 11 showed that the 
mechanism is most likely simple alkyl exchange of the methyl groups 



H 3° > 

H 3 C * 



III ^' PPh 3 



+ Au 1 
I 
PPh, 



H 3° 
H3C 



PPh. 



Au 



III 



CH 



Au J 



PPh. 



CH 



^ Au 111 



,CH 



CH. 



I 
J + Au 1 



PPh. 



PP] 




Alkyl Isomerizat ion 



(CH 3 ) 2 ( (CH 3 ) 3 C-) AuPPh 3 was found to isomerize 
to the iso-butyldime thyl complex in diethyl ether a 
temperature. 12 The kinetics of the isomer izat ion w 
first order in (CH 3 ) 2 ( (CH 3 ) 3 C-) AuPPh 3 . The first o 
constant of this reaction was found to decrease wit 
amounts of PPh 3 . An associative mechanism involvin 
PPh 3 was ruled out by X P NMR studies. A dissociat 
was proposed to explain the reduction in the rate b 
It appears that the isomerization is primarily due 
factors since an isopropylgold (III ) compound showed 
rearrangement to an n-propyl compound until thermal 
occurred. Analysis of the decomposition products i 
presence of an n-propyl compound. 



spontaneously 
t room 

ere found to be 
rder rate 
h increasing 
g the added 
ive mechanism 
y added PPh 3 . 
to steric 
no evidence of 
decomposition 
ndicated the 




Reductive elimination from trialkyl gold species results in 
the coupling of two of the alkyl groups rather than dispropor- 
tionation. 11 Reductive elimination in nonpolar solvents show 



CH 3 

I 
R-Au-PPh 3 

I 
CH 3 



j — > RCH 3 + CH 3 AuPPh 3 
\h> R(-H) + CH„ + CH 3 AuPPh 9 



the participation of an int ermolecular pathway while polar solvents 
show exclusively an intramolecular process. The intramolecular 
pathway can be interpreted by a mechansim that encompasses a three 
coordinate trialkyl species. Theoretical studies 1 ** indicate that the 
most symmetrical (C 3 h) intermediate is Jahn-Teller active and thus 
favors a distortion to T and Y shaped intermediates. 



R R 




1 





The T and Y shaped geometries each show three equivalent energy 
conformations that are lower in energy than the Ca^ geometry. 
The T shaped geometries are the energy minima whereas the Y shaped 
geometries are saddle points that serve as exit channels for the 
eeupled alkane . 

Conclus ion 

The systematic study of the mechanism of reductive coupling 
alkyl groups on organogold compounds has lead to some generalizations 
concerning the processes involved. Oxidative addition of alkyl 
halides to CH 3 Au^L was found to be dependent upon L. The rate 
of addition was found to be L = P (CH 3 ) 3 >PPh 3 . Reductive 
elimination is observed through a three coordinate species that by 
theoretical studies is shown to be distorted from a highly 
symmetrical intermediate to a energetically more favorable T or 
Y shape. This distortion allows the appropriate orbitals of the 
methyl groups to interact forming the carbon-carbon bond of the 
elimination product. 



References 



2. 
3. 
4, 
5, 
6, 
7, 
8, 
9 

10 
11 
12 

13 

,14 



B. Armer, H. Schmidbaur, Angew 
(1970) . 

H. Schmidbaur, Angew. Chem. in 
E. Frankland and D. Duppa, J. 
W. Pope and C. Gibson, J. Chem 
L. Woods and H. Gilman, J. Ame 
G. Coates and C. Parkin, J. Ch 
A Tamaki and J. Kochi, J. Orga 
A. Tamaki and J. Kochi, J. Org 
A. Johnson and R. Puddephatt, 
(1975). 

A. Tamaki and J. Kochi, J. Che 

Tobias , J . Orga 
Magennis, J. Koc 



Chem. internat. Edit., 9, 101 



G. Rice and R 
A . Tamaki , S . 
6140 (1974). 
Cf. F. Basolo 
2d ed , Wiley , 
S . Komiya , T . 



and R. Pearson, 
New York, N.Y. , 
Albright, R. Hof 



ternat . Edit . , ' 
Chem . Soc . (Lon 
. Soc. (London) 
r . Chem . Soc . , 
em. Soc. (Londo 
nometal. Chem., 
anometal. Chem. 
J. Organometal. 

m . Soc . Dal ton 
nometal . Chem. , 
hi , J . Amer . Ch 

"Mechansims of 
1957, p. 375. 
f man , J . Kochi , 



15, 728 (1976). 
don) 17, 29 (1864). 

91, 2061 (1907). 
70, 550 (1948). 
n) 1963, 421. 

40, C81 (1972). 
, 6±, 411 (1974) . 
Chem. , 8_5, 115 

Trans., 2620 (1973). 

86, C37 (1975). 
em. Soc . , 96 , 

Inorganic Reactions," 

J. Amer. Chem. 



Soc, 98, 7255 (1976). 



21 

P- ORBITAL CHEMISTRY 
Jeorge M. St. George November 10, 1977 

The f-orbitals in lanthanide and actinide elements are of roughly 
;he same energy as the outer s-, p-, and d-orbitals of these elements, 
md so are energetically available for chemical reactivity. The size 
Hid spatial distribution of the f-orbitals can provide the rare-earth 
elements with a chemistry not found in d-orbital transition metals. 

i 

It was originally thought that, while energetically feasible, 
:ovalent bonding to rare earths was impossible because of the small 
radial extension of the f-orbitals and that electrostatics was the 
lorn i riant theme of lanthanide and actinide chemistry. Early experi- 
nental evidence supported this contention: homoleptic rare-earth 
ilkyls, where extensive covalency is expected, were for a long time 
msynthesizable; synthesis of halogen-, oxygen-, and nitrogen-donor 
complexes, on the other hand, were made quite easily; the electronic 
spectra of these latter complexes were almost identical to those of 
bhe free ions--calculation of the nephelauxetic parameter, 3, indicates 
Less than 5% covalency in these complexes; 1 > 2 and the magnetic proper- 
ties of these complexes are also very similar to those of the free ions. 

More recently, several organo-lanthanide and -actinide compounds 
have been made. The earliest were ti 5 -C 5 H5 compounds, followed by 
n-C 8 H a , n 3 -C 3 H5, and n 6 -C 6 H 6 compounds. The electronic spectra of 
Lanthanide and actinide tricylopentadienides indicate minimal covalency 
(5/j) in these compounds, 3 but still greater than in non-carbon-donor 
compounds. 1 H-nmr studies of U(C 5 H 5 T4, 4 ^ 5 Th(C 5 H 5 )4, 5 and U(ti 3 -C 3 H 5 )4 6 
indicate that dipolar contributions are most important in the isotropic 
proton shifts, indicating little covalency in these complexes. 1 H-nmr 
studies of cyclooctatetraenyl compounds, however, indicate not only 
strong covalent interactions but also that these interactions are ligand rr 
to metal-f charge transfer in character. 7 

Early attempts at synthesis of homoleptic alkyl complexes led to 
pyrophoric, uncharacterized complexes. By blocking off most of the 
metal coordination spheres with cyclopentadienyl rings, compounds of 
the form (C 5 H 5 ) MR were made; 8 » 9 » 10 ' 11 these compounds have high 
thermal stability, though they are highly sensitive to oxygen and 
moisture. Nmr studies indicate high covalency in the M-R bond. In 
lanthanide complexes of this type, a ligand-to-metal charge transfer 
band is seen in the near UV; hypersensitivity is seen in the f-f 
transitions of certain compounds of this type, indicating covalency 
in the bonding. X1 

The decomposition of (C 5 H 5 )3ThR has been studied in detail; 9 
decomposition occurs by intramolecular abstraction of a proton from a 
C 5 H 5 ring by the alkyl group. 



2(C 5 H 5 )3ThR -> 2RH + (CsHs^Th^ N ^ v -Th(C 5 H 5 )2 





22 

A "carbene" intermediate, nV=Th(C 5 H 5 ) 2 > has been proposed for this 

reaction. Although no stable carbene complex of a rare-earth element 
has been synthesized, a recent angular-overlap-model study has demon- 
strated the possibility of n-bonding in this manner by f-orbitals. 12 

With the exception of (0 4 La)~ and (0 4 Pr)~, homoleptic rare- earth 
alkyl complexes were uncharacterized until quite recently. In situ 
reaction of UC1 4 ,wlth four equivalents of RLi led to alkane and aikene 
thermolysis products via a proposed UR4 intermediate. 13 Recent evi- 
dence indicates that the intermediate may be of the form (UR 6 )~> as 
isolable compounds of this type have been synthesized, as have (ur 8 ) 3 ~ 
complexes. 14 The octaalkyluranate(v) complexes have remarkable thermal 
stability and appear to be coordinatively saturated. 

As well as the -recent appearance of new and interesting lanthanide 
and actinide complexes, °> 1X > lfe there has been great interest in finding 
new uses for longer-known complexes utilizing the peculiar coordination 
geometries of f-orbital elements. 

While the rare earths have been a subject of study for many years, 
it was not until the late 196o*s and the discovery of isolable organo- 
metallic complexes of these elements that intensive study was under- 
taken; and much remains to be learned about these elements and their 
compounds. Indications of covalency in these complexes supports the 
importance of the f-orbitals (as opposed to other effects) in the 
unique chemistry of rare-earth complexes. There is great promise of 
exciting discoveries, both theoretical and synthetic, in this field. 

REFERENCES : 

1. S. P. Tandon and P. C. Mehta, J. Chem. Phys., 52, 4896 (1970). 

2. Tandon and Mehta, ibid., 5417. 

3. L. J. Nugent, et al . , J. Organoraet. Chem., 27, 365 (l97l). 

4. R. von Ammon ancl B. Kanellakopulos, Chem. Phys. Lett., _4, 
553 (1970). 

5. von Ammon and Kanellakopulos, ibid . , 2, 513 (1968). 

6. N. Paladino, et al., ibid ., 5,~T5~(l970) . 

7. A. Streitweiser, Jr., et al., J. Amer. Chem. Soc, 95 , 
7343 (1971). 

8. T. J. Marks, et al., J. Amer. Chem. Soc, 95, 5529 (1973). 

9. T. J. Marks, ibid . , 223. 

10. M. Tsutsui and N. M. Ely, J. Amer. Chem. Soc., 97, 1280 (1975). 

11. M. Tsutsui, et al. , Accts. Chem. Res., 9, 217 (1976). 

12. K. D. Warren, Inorg. Chem., l6, 2008 (1977). 

13. T. J. Marks and A. M. Seyam, J. Organomet. Chem., 67, 6l (1974). 

14. E. R. Sigurdson and G. Wilkinson, J. Chem. Soc. DaTEon, 1977 , 812. 

15. J. Holton, et al., J. Chem. Soc. Chem. Comra., 1976 , 480. 

General References: 

Cotton, F. A., and Wilkinson, G. , Advanced Inorganic Chemistry, 

John Wiley and Sons, 1972. 
'H. G. Friedman, Jr., et. al., J. Chem. Educ, 4l, 354 (1964). 
Clifford Becker, ibid., 358. 



23 



THEORY AND APPLICATIONS OF EXTENDED X-RAY ABSORPTION FINE 
STRUCTURE ANALYSIS (EXAFS) 

Paul Young November 17, 1977 

In recent years, analysis of the irregular oscillations in the 
atomic x-ray absorption coefficient (EXAFS) has provided valuable 
structural information on systems where conventional x-ray 
diffraction methods are inapplicable. Since EXAFS can be observed 
for gasses, solutions, and amorphous solids, exciting advances 
have been made involving these forms of matter. 

Although EXAFS has been known since the 1920's, 1 no satisfactory 
and useful theory was presented until 1971. 2 Present theories, 
which are used to obtain interatomic distances, explain EXAFS as 
a backseat tering interference effect on the final state in the 
photoelectric absorption process. For the purposes of EXAFS, 
the ejected pho toelec t ron can be considered as an expanding spherical 
wave of wave vector k, k=/Z (fi-E ) , where 0, is the x-ray energy and 
E is the threshold energy. 3 This final state will be scattered and 
phase shifted by the neighbors, and the backseat tered wave will 
constructively or destructively interfere with itself at the 
absorbing atom, thus enhancing or reducing the absorption 
probab ili ty . * Since the backseat tering factor, f, and the phase 
shift, n , are functions of k, the final effect on the absorption 
coefficient, X (k) > i- s not simple. 5 Because the phase shift and 
wave vector are not known, problems of interpretation have plagued 
investigators. 6 Several methods have been published to determine 
H and E , 6 ' 7 ' 8 but in practice, model compounds of known 
structure are used to empirically determine atom-pair phase shifts 
and to guide judicious truncation of data to minimize the effect 
of choice of E . The Fourier transform of the data is then 
obtained, giving the radial distribution of scatterers. 9 ' 10 
Least squares fitting of the EXAFS parameters to the data has also 
been accomplished. 11 

Experimentally, EXAFS has experienced a renaissance with the 
completion of the high intensity x-ray source at the Stanford 
Linear Accelerator Center. 12 This beam is 10 b times brighter than 
conventional x-ray tubes, and provides a continuous spectrum 
from 4000A down to 0.3A. A crystal monochroma to r is used to 
select energy and obtain the EXAFS spectrum. 

Many systems have been investigated with EXAFS. The first 
Fourier transform studies of the EXAFS method compared the known 
structure of crystalline germanium with that of amorphous germanium 
and found smearing due to distortion of bond angles. 2C In a more 
recent example, the coordination of copper in the semiconductor 
Cu^(Asq /Sen f ^i-x has been shown to be quite similar to that in 
crystalline CuAsSe 2 . 13 



24 



EXAF 
as well . 
and of th 
impressiv 
of in viv 
iron(III) 
and hydro 
Finally , 
the combi 
s true ture 
coo r dina t 
Ru-0 dist 



S is a 

The d 

ree su 

e exam 

o mang 

and n 

xide b 

EXAFS 

nation 

of su 

es to 

a n c e a 



use f 
e tec t 
lfur 
pies . 
anes e 
ickel 
ridge 
of hi 

of E 
r f ace 
a Cu- 
s in 



ul tool for study of bioinorganic molecules 
ion of four equivalent Fe-S bonds in rubredoxin 
atoms to molybdenum 1 " 5 in nitrogenase are 
An EXAFS spectrum has even been taken 
of a leaf. The coordination spheres of 
(II) in aqueous solution have been determined, 
d dimers of iron(III) have been found. 17 ' 18 
ghly dispersed supported metal particles and 
XAFS and Auger techniques have probed the 
s. Oneexample is the discovery that dioxygen 
Ru catalyst at a Ru 3 group with the same 
Ru0 2 . 19 



1 u 



In summary, EXAFS, which is dependent on backs cat ter ing by 
the local environment of the absorbing atom, is one of the most 
powerful and exciting structural techniques in many years. 
Detailed studies -o f previously unknown structures have advanced 
markedly, and further advances will tantalize chemists for years 
to come . 



References 



1, 
2, 

3, 

4, 

5 

6 

7 
8 
9 

10 
11 
12 
13 

14 

15 
16 
17 
18 

19 



Am ., Apr . 

Pendry , Ph 
Eisenber g 



A. H. Compton, S. K. Alliso 
p 662, D. Van Nostrand Co. P 
D. E. Sayers, E. A. Stern, 
27 , 1204 , (1971) . b) Phys 
c) Phys. Rev. B, 11, 4836, 

Lagarde, Phys. Rev. 3, 1 

A. Stern, Sci 

A . Lee , J . B . 

M . Kincaid , P 
(1975) . 

C. A. Ashley, S. Doniach, P 
P. A. Lee, G. Beni, Phys. R 

Citrin, P. Eisenberge 
3_6, 1346 (1976) . 
Gurman, J. B. Pendry, 
Te o , et al . , J. Amer. 

Science, 190 1074 (1975). 

S. Hunter, A. Bienenstock, 

D. E. Sayers, E. A. Stern, 
64 , 427 (1976) . 

E. I. Stiefel, Inorganic Se 
J. Jaklevid et al . , Solid S 

F. W. Lytle, J. Chem. Phys. 
A. H. Reis, A.C.S. National 

(1977) . 

Science, 197 , 34 (1977) . 



P . 

E. 
P . 
B. 



P . H 
Lett. 
S . J 

B.-K 



n , X-Rays in Theory and Experiment , 
rinceton, N.J. (1935). 
F. W. Lytle; a) Phys. Rev. Lett., 
. Rev. B, 11, 4825, (1975) . 

(1975) . 
4, 741 (1976). 
1976, 96 (1976). 
ys. Rev. B, 11, 2795 (1975). 
er, Phys. Rev. Lett., 3_4 , 1361 

hys. Rev. B, 1_1, 1279 (1975). 

ev. B, 15, 2862 (1977) . 

r, B. M. Kincaid, Phys. Rev. 



Solid State Coram., 2_0_ , 287 
Chem. Soc, 99, 3856 (1977) 



(1976) 



SSRP Report 76102 (1976) . 

J. R. Herriott, J. Chem. Phys., 

minar, Oct. 25 (1977). 
tate Co mm., 2_3, 679 (1977). 
, 67_, 473 (1977) . 
Meeting, Inor. B .> #193, Chicago 



MECHANISTIC STUDIES OF THE OLEFIN METATHESIS REACTION 



25 



Bruce Calvert 



December 1, 1977 



Within three years 
of olefins (1964) the p 
A wide variety of homog 
reported to catalyze th 
olefin types. Followin 
bonds were broken durin 
mechanisms appeared whi 
t ransalky lidena t ion . T 
" quasicy clobut ane " or "t 
cyclopentane. More rec 
proceeds via a metal c 
mechanistic scheme impl 
1970 by Chauvin and Her 
molecular weight polyal 
metathesis of cycloolef 
been conducted which su 
cyclobutane as the key 
mechanistic scheme inco 
generally accepted and 



of the fi 
rocess had 
eneous and 
e me t a thes 
g the disc 
g the cour 
ch shared 
he transit 
et rame thy 1 
ent s tudie 
arbene- ini 
icating me 
r is son to 
kenamers a 
ins . 3 Ver 
ppo r t the 
step in th 
rpora t ing 
is p resent 



rst reported catalytic metathesis 
become industrially significant. 1 
heterogeneous catalysts have been 
is of an equally diverse set of 
overy that the olefin double 
se of the reaction, 2 several 
the common feature of "pairwise" 
ion state was hypothesized to be 
ene" metal complexes or a metallo- 
s suggest that the reaction 
tiated chain process. The first 
tal carbenes was proposed in 

explain the formation of high 
t low conversions during the 
y recently, model studies have 
formation of a puckered metallo- 
e metathesis of olefins. A 
these features has become 
ed below . 



M = C 



/ 



R 



b' 



M 



a-C 



C-R 
C — 



C 



C 

c / \ 

a 7 x b 



Th 
"first- 
the p re 
was ope 
cross m 
of iso t 
that t h 
for the 
quenchi 
con ta in 
isolate 
studies 
first r 
number 



e forma 
formed" 
sence o 
rat ing 
etathes 
opicall 
e chain 

me tal 
ng exp e 
ing ele 
d under 

The 
epo r t ed 
o f mod e 



t ion 

pro 
f se 
in t 
is o 
y la 

mec 
carb 
rime 
c t ro 

met 
i so 1 

by 
llin 



o f high 
due t s of 
squi-oli 
he ca t al 
f acy cl i 
belled a 
han i sm w 
ene mech 
n ts with 
n-rich o 
a thes is 
a t ion of 
Casey 
g e xpe r i 



mole 
the 
gome 
ys t 
c an 

, 03- 

as o 

anis 

Mic 

lef i 

cond 

non 

and 

men t 



cul a 
met 

6 

r s 

sy s t 
d cy 
d ien 
pera 
m ha 
hael 
ns i 
i t io 
-he t 

by 
s . 



r we 
a the 
sugg 
ems 
cloo 

8 

es 
ting 

S CO 

ace 

n wh 
1 
ns , 

e roa 

Schr 

In a 



ight 
sis 
es t e 
stud 
lefi 
p rov 

in 
me r 
ep to 
i ch 
1 an 
torn 
ock , 
t le 



poly 
of cy 
d tha 
ied . 
ns 7 a 
ided 
these 
ecen t 

1 o 

rs , 

me tal 

d fro 

stabi 
1 2b . 

as t o 



alkenamer s as 

cloolef ins 5 and 

t a chain mechanism 

However, only the 
nd the metathesis 
unequivocal evidence 

systems. Support 
ly also from 

f rom sy s t ems 

ca rbenes can b c 
m numerous model 
lized metal carbenes, 
nitiated a large 
ne case , a me ta 1 



26 



carbene has been shown to be an active catalyst for olefin 
me ta t hes is . 

In an elegantly designed labelling study Grubbs has elucidated 
the mode of initiation for two different homogeneous catalyst 
systems. 9 A parallel between metathesis activity and extent of 
formation of active catalyst in the initiation step was shown 
and the suggested dimerization of me tal-me thy lene species would 
provide a chain termination mechanism. 



Because of the range of catalysts for the olefin metathesis 
reaction and the variety of olefin types studied, it is entirely 
possible that no single mechanistic scheme will prove general. 
However, the suggestion that the olefin metathesis reaction is 
initiated by a metal carbene complex and proceeds through a 
met allocy clobut ane* intermediate allows rationalization of all 
the presently available data, including the details of selectivity 
during metathesis 



1^)15 



Ref er ences 



9 
10 

11 

12a 



13 
14 

15 



R. J. Haines, G. J. Leigh, Chem. Soc . Rev., 5_, 155 (1975). 

N. Calderon, E. A. Ofstead, J. P. Ward, W. A. Judy, and 

K. W. Scott, J. Am. Chem. Soc, 9_0, 4133 (1968), and references 

therein; J. C. Mol, J. A. Moulijn, and C. Boelhouwer, 

J. C. S. Chem. Coram., 633 (1968). 

J. L. Herisson and Y. Chauvin, Makromol. Chem., 141, 161 (1970) 

C. P. Casey, L. D. Albin, and T. J. Burkhardt, J. Am. Chem. 

Soc. , 99, 2533 (1977) . 

5, 127 (1972). 



N. Calderon, Ace. Chem. Res 

W. J. Kelly, N. Calderon, J 

(1975) . 

T. J. Katz, J. McGinnis, J. 
Grubbs, D. D. Carr, C 
98^, 3478 (1976); T. J 
_9_8, 2519 (1976). 
Grubbs and C. R. Hoppin, J 
Gassman and T. H. Johnson, 



R. H. 

Soc . , 

Soc., 

R. H. 

P . G. 

(1976) . 

D . J . Card in , 

927 (1972) . 

C . P . Casey , T 

(1973) ; b. R. 

(1974) . 

T. J. Katz and 
C . P . Casey , H 
98, 608 (1976) 



Macromol. Sci.-Chem., A9^, 911 

Am. Chem. Soc, 9_9, 1903 (1977). 
Hoppin, P. L. Bruk, J. Am. Chem. 
Katz, R. Rothchild, J. Am. Chem. 



C. S. Chem. Coram., 634 (1977) 
J. Am. Chem. Soc, 98, 6058 



M 



J. Doyle, M. F. Lappert, J. C. S. Chem. Comm 



J 

R. 

W. 



Burkhard t , J . 
Schrock, J. Am 



Am. Chem. Soc, 9J5 , 5833 
Chem. Soc , 96, 6796 



H. Hersh, Tet. Lett., 585 (1977). 

E. Tuinstra, M. C. Saeman, J. Am. Chem 

and references therein. 
See for example, J. L. Bilhou, J. M. Basset, R. Mutin, 
W. F. Graydon, J. Am. Chem. Soc, 9_9, 4083 (1977). 



Soc . , 



27 

EXPERIMENTAL APPLICATIONS OF THE ANGULAR OVERLAP MODEL 
Kenneth Leslie December 9, 1977 

The spectral and magnetic properties of transition metal 
complexes can to a large extent be explained in terms of the 
d electron configurations of the metal. Crystal field, ligand 
field, and molecular orbital calculations have all been used 
to explain these properties. Methods have been developed which 
try to combine the principles of semi-empirical molecular 
orbital calculations with chemical intuition. An approach 
of this type is the angular overlap model (AOM), "^ the name 
being derived from the fact that the squares of the overlap 
intergrals between the metal d orbitals and the appropriate 
ligand orbitals are proportional to the destabilization energy 
of the d orbitals in the complex. 

The AOM parameters can be used to describe the energies of 
the d orbitals in terms of a , it, and 5 interactions. The 
parameters arise from a perturbation theory treatment coupled 
with the Wolf sberg-Helmholz approximation. This allows the 
destabilization energy of a d orbital to be written as 

2 



ml 

\ h m" h : 



E -l-u^J s£ L 



where H y and H M are the coulomb integrals for the ligand and the 
metal, and Sf-jL is the overlap, integral . By factoring Sj^l into 
radial (Sjvjl") and angular ( F^ ) parts, the energy is now written 



as 



5& L ) (*) 2 (-/)' 




where e^ is defined as the AOM radial parameter with 

A = a, it, 6 for each type of orbital interaction. Ff , with 

I denoting the azimuthal quantum number, is the angular overlap 

integral and is obtained from matrices derived for p, d, and f 

orbitals. 7 5 8 For a metal complex with N ligands , it is assumed 

that the contributions from each ligand are additive giving 




The total angular overlap effect is found by summing the Fa 
values obtained for each ligand having coordinates (0]<, <$>y) . 
Consequently, each d" level can be described using a combination 
of angular coefficients and radial parameters of the form 
Ae a + Be^ + Ce<5 . 

The AOM has been applied to a wide variety of chemical 
systems and has yielded many types of chemical correlations. 



A series of trans-tetraamine and trans-tetrakis (pyridine) 
chromiumdll ) complexes has been examined using the AOM. 9 
Solution absorption spectra and a transferability assumption for 
parameters were used to construct a two-dimensional spectro- 
chemical series of ligands for chromiumdll ) . The tt parameter 
for the chromiumdll) to pyridine bond was found to be negative, 
indicating the presence of 7T-back bonding. The single crystal 
linearly polarized electronic spectra of tetragonal Ni(en)2 - 
(NCS)2 and Ni(en) 2 (NO2 ) 2 were used to derive e a and e^ 
parameters.^ 'These were used along with other nickel amine 
AOM values to determine a linear relationship between e^ and 
nickel-nitrogen distance. Correlations of this type are valuable 
in interpreting other systems as evidenced by Purcell's study of 
the isomers Fe(phen) 2 (NCBH3 ) 2 and Fe (phen) 2 ( CNBH3 ) 2 . 11 Apparent 
discrepancies among the IR, magnetic, and Mossbuaer data were 
explained using AOM correlations. 

The AOM has also found use as a means for determining the 

relative energies of the d orbitals in a number of copper(II) 

complexes . 12-15 /\ study of some copper(II) acetylacetonate 

complexes has verified the orbital energy sequence suggested 

by Bel ford. Likewise, the energies of the electronic states 

of Co(salen), Co( salen)dimer , and Co(salen)py have been 

calculated^' using the AOM and they satisfactorily account for 

the observed electronic and EPR spectra. Gerloch has obtained 

the same type of success using the AOM to parameterize his ligand 

field method for calculating optical and EPR spectra and magnetic 
susceptibilities. -^-20 

The angular overlap model is beginning to have widespread 
application in the correlation and rationalization of experimental 
results on transition metal systems. Due to the chemical 
significance of its parameters, the AOM should continue to gain 
popularity among inorganic chemists. 

Re ferences 

1. C. E. Schaffer and C. K. Jorgensen, Mol . Phys . 9_, 401 (1965). 

2. C. E. Schaffer, Struct, and Bonding 5, 6 8 (19 68). 

3. C. E. Schaffer, Pure Appl. Chem. 2^4_, 361 (1970). 

4. C. E. Schaffer, Struct, and Bonding 14, 69 (1973). 

5. E. Larsen and G. N. LaMar , J. Chem. Ed. 51, 633 (1974). 

6. M. Wolfsberg and L. Helmholz, J. Chem. Phys. _2_0, 837 (1952). 

7. M. Gerloch and R. C. Slade, "Ligand Field Parameters," 
Cambridge University Press, London, 197 3. 

8. W. Smith and D. W. Clark, Rev. Roum. Chim. 20_, 1243 (1975). 

9. C. E. Schaffer, et. al. , Inorg. Chem. 15, 1399 (1976). 

10. I. Bertini, et. al. , Inorg. Chem. 15, 203 (1976). 

11. K. F. Purcell, et . al. , Inorg. Chem. 16, 1708 (1977). 

12. F. J. C. RossottT, et. al. J. C. S. Dalton, 1509 (1972). 

13. A. B. P. Lever, Coord. Chem. Rev. 3_> 119 (1968). 

14. D. W. Smith, J. Chem. Soc. A, 1209 (1971). 

15. A. Bencini and D. Gatteschi, Inorg. Chem. 1_6, 1994 (1977). 

16. M. A. Hitchman, J. C. S. Faraday II, 846 (1972). 

17. M. A. Hitchman, Inorg. Chem. 16, 1985 (1977). 

18. D. A. Cruse and M. Gerloch, J. C. S. Dalton, 152 (1977). 

19. D. A. Cruse and M. Gerloch, J. C. S. Dalton, 1613 (1977). 

20. M. Gerlochand R. F. McMeeking, J. C. S. Dalton, 2443 (1975). 



27 

CHEMISTRY OF THE EXCITED STATE OF 
TRIS(2,2'-BIPYRIDINE)RUTHENIUM(ll) 

Nancy P. Forbus December 15, 1977 

Since the first reported use 1 of tris( 2, 2' -bipyridine ) - 
ruthenium(ll) , Ru(bipy) 3 2 , as a photosensitizer in the aquation 
of PtCl 4 2 , a great deal of work has been done on the luminescent 
quenching of Ru(bipy) 3 2 . Several features 2 of Ru(bipy) 3 2 make it 
especially attractive for study; these include water solubility, 
strong visible absorption (e~l4000 at 450 nm) , photostability, 
a low-lying (17.<^ kk) excited state, and strong luminescence with a 
lifetime of approximately 0.7 microseconds in aqueous solutions at 
room temperature. 

Although the exact nature of the reactive excited state has 
not been completely resolved, studies indicate that it is a manifold 
of three thermally equilibrated metal-ligand charge transfer states 
of Aij E, and A 2 symmetry. 3 There is general agreement that the 
emitting state is predominately triplet in character, although the 
large amount of spin-orbit coupling present makes difficult an exact 
assignment to the spin multiplicity. 

Energy transfer is one important mechanism 4 for quenching in 
Ru(bipy) 3 2 . Two studies that illustrate unambiguous energy transfer 
in fluid media involve quenching by Cr(CN) 6 3 ~ 5 and by a series of 
organic compounds 6 known to be efficient triplet energy acceptors. 

Since there is considerable charge separation in the luminescent 
excited state, 7 Ru(bipy) 3 2 should have both oxidizing and reducing 
sites and electron transfer quenching, both oxidative and reductive, 
could be expected. Oxidative quenching of Ru(bipy) 3 2 has been ex- 
tensively studied, but direct observation of the electron transfer 
products is made difficult by the fact that the Ru(bipy) 3 3 formed 
in the quenching process is a strong oxidizing agent. 7 As a result, 
the initial quenching reaction is followed by a rapid dark thermal 
reaction to regenerate Ru(bipy) 3 2 and the oxidized form of the 
quencher. 8 The mechanism of the quenching by Co(NH 3 ) 5 X (X=NH 3 , H 2 0, 
CI , Br ) resulting in the redox decomposition of the Co(IIl) complex 
has been the subject of some controversy, 9-11 but electron transfer 
is presently favored. 12 Flash photolysis 13 and steady state 14 studies 
of quenching by Fe(H 2 0) 6 3 indicate that this species also quenches 
by an electron transfer mechanism. Oxidative quenching of Ru(bipy) 3 2 
has been observed for various organic quenchers. As the reduction 
potentials of members of a series of nitrobenzene derivatives are 
varied from -0.525 V to -1.203 V, the observed quenching rate constant 
decreases by four orders of magnitude. 15 Two other organic quenchers 
which cause oxidative quenching are 1, 1 ' -dimethyl-4 , 4 ' -bipyridine 
(paraquat, P 2 ) and trans-1, 2-bis(N-methyl-4-pyridyl)ethylene (trans- 
4,4' -BPE). 13 That electron transfer quenching is observed for trans- 
4,4'-BPE is significant because this quencher has a low- lying triplet 
state that should be readily accessible to Ru(bipy) 3 2 for energy 
transfer quenching. 



28 



2 + 



The first suggestion that reductive quenching of Ru(bipy) 3 
occurs was made by Creutz and Sutin 16 who observed that Ru(NH 3 ) 6 2f , 
S 2 4 2 ~> Fe(CN) 6 4 ~, and Eu(lll) quenched the Ru(bipy) 3 2 luminescence 
with rates very similar to the rates at which these species were 
known to reduce horse heart f erricytochrome c. Balzani and co- 
workers 17 found evidence for the formation of Mo(CN) 8 3 ~ and Os(CN) 6 3 ~ 
when Mo(CN) 8 4 and Os(CN) 6 4 ~, respectively, were used as quenchers. 
Further evidence for reductive quenching comes from studies using 
organic quenchers. When N,N,N ! ,N' -tetramethyl-p-phenylenediamine 
(TMPD) and 10-methylphenothiazine (10-MPTH) were used as quenchers, 
spectral evidence for formation of TMPD and 10-MPTK was obtained. 18 
The absorption spectrum of Ru(bipy) 3 + calculated from flash photolysis 
studies of the TMPD system agreed well with that obtained from electro- 
chemically generated Ru(bipy) 3 . 

The observed photochemistry of Ru(bipy) 3 2 suggests several 
applications in converting light energy to more useful forms. In a 
study using paraquat (P 2 ) as a quencher in the presence of triphenyl- 
amine, Ru(bipy) 3 ^ has been observed to drive the reaction P 2 + 
NPh 3 -> P + NPh 3 in the nonspontaneous direction with conversion of 
69% of the excited state energy of Ru(bipy) 3 2 into the chemical 
energy of the P 2 -NPh 3 system. 19 Sutin has constructed a photogalvanic 
cell, in which the Ru(bipy) 3 2 -Fe(lll) system is observed to generate 
a potential upon irradiation of half of t£e cell with light of a wave- 
length appropriate to produce Ru(bipy) 3 2 . 2 ° Sutin has suggested 
a design for a cell containing only Ru(bipy) 3 2 in basic solution with 
an n-type semiconducting electrode and a Pt electrode that should 
theoretically be able to reduce water to H 2 and 2 . 21 However, con- 
struction of a working model has not been reported to date. Whitten's 
report 22 of surfactant derivatives of Ru(bipy) 3 2 causing decompo- 
sition of water was later found to be irreproducible 23 upon rigorous 
purification of the surfactants; it was suggested that some impurity 
was the reactive species, but attempts to identify that species have 
so far been unsuccessful. 

These studies have shown that the chemistry of excited state 
Ru(bipy) 3 2 is quite complex. Perhaps by judicious substitution on 
the bipyridyl ligand, 14 one could control emission lifetimes and 
reduction potentials of Ru(bipy) 3 2 compounds, making possible the 
design of Ru(bipy) 3 2 compounds that will exhibit desirable excited 
state properties. 



29 

1. J. N. Demas and A W. Adamson, J. Am. Chem. Soc, 93, l800 (1971). 

2. J. N. Demas and A. W. Adamson, J. Am. Chem. Soc, 95, 5159 (1973). 

3. G. D. Hager and G. A. Crosby, J. Am. Chem. Soc., 97, 7031 (1975). 

4. J. N. Demas and J. W. Addington, J. Am. Chem. Soc, 98, 5800 (1976). 

5. N. Sabbatini and V. Balzani, J. Am. Chem. Soc, 9k, 7587 (1972). 

6. M. Wrighton and J. Markham, J. Phys. Chem., 77, 3042 (1973). 

7. C P. Anderson, D. J. Salmon, T. J. Meyer, and R. C. Young, 
J. Am. Chem. Soc, 99, 198l (1977). 

8. C. R. Bock, T. J. Meyer, and D. G. Whitten, J. Am. Chem. Soc, 
96, 4710, (1974). 

9. H. D. Gafney and A. W. Adamson, J. Am. Chem. Soc, 9^, 8238 (1972). 

10. G. Navon and N. Sutin, Inorg. Chem., L3, 2159 (197477 

11. (a) P. Natarajan and J. F. Endicott, J. Phys. Chem., 77, 971 (1973). 

12. (b) P. Natarajan and J. F. Endicott, J. Phys. Chem., 77, 1823 (1973) 
V. Balazni, L. Moggi, M. F. Bolletta, and G. S. Laurence, 

Coord. Chem. Revs., 15, 321 (1975). 

13. C. R. Bock, TJ J. Meyer, and D. G. Whitten, J. Am. Chem. Soc, 

96, 4710 (1974). 

14. C. T. Lin, W. Bottcher, M. Chou, C. Creutz, and N. Sutin, 
J. Am. Chem. Soc, 98, 6536 (1976). 

15. C. R. Bock, T. J. Meyer, and D. G. Whitten, J. Am. Chem. Soc, 

97, 2909 (1975). 

16. C. Creutz and N. Sutin, Inorg. Chem., 15, 496 (1976). 

17. A. Juris, M. T. Gandolfi, M. F. Manfrin, and V. Balzani, 
J. Am. Chem. Soc, 98, 1047 (1976). 

18. C. P. Anderson, D. J. Salmon, T. J. Meyers, and R. C. Young, 
J. Am. Chem. Soc, 99, 1980 (1977). 

19. R. C. Young, T. J. Meyer, and D. G. Whitten, J. Am. Chem. Soc, 
97, 4782 (1975). 

20. C. T. Lin and N. Sutin, J. Phys. Chem., 80, 97 (1976). 

21. C. Creutz and N. Sutin, Proc. Nat. Acad. Sci. USA, 72, 2858 (1975). 

22. G. Sprintschnik, H. W. Sprintschnik, P. P. Kirsch, and 
D. G. Whitten, J. Am. Chem. Soc, 98, 2337 (1976). 

23. G. Sprintschnik, H. W. Sprintschnik, P. P. Kirsch, and 
D. G. Whitten, J. Am. Chem. Soc, 99, 4947 (1977). 



30 

THE PARTIAL MOSSBAUER PARAMETER AND THE TRANS INFLUENCE 

IN LINEAR GOLD (I ) COMPOUNDS 

Michael J. Desmond January 2k, 1978 

The Mossbauer effect 1 has been observed for over thirty iso- 
topes. The Mossbauer observables are the center shift (c.S.) which 
results from the variation in nuclear electron density and the 
quadrupole splitting which results from the electric field gradient 
at the nucleus. For low spin Fe(ll) and for Sn(iv) systems, which 
have no contribution to the field gradient from asymmetric valence 
electron distribution, ligand contributions to the spectra have 
been successfully quantified through use of the additivity model. 2 * 3 
The resulting ligand partial quadrupole splitting (p.q.s.) and 
partial center shift (p.c.s.) values have been used to give an indi- 
cation of the modes of bonding and elucidate the structures of vari- 
ous compounds. A study of methyl tin iodides, 4 however, exposed the 
shortcomings of the electrostatic model for the additivity of ligand 
effects. The results were interpreted on the basis of the p-orbital 
imbalance involved in the bonding of methyl groups and iodide ligands. 
Furthermore, the additivity model does not successfully include rr 
bonding effects and significant deviations from predicted quadrupole 
splitting values have been observed for Fe(ll) compounds containing 
strong n acceptor ligands. 5 

Recent M5ssbauer investigations 6 * 7 of l97 Au(l) (d 10 ) compounds 
have applied the additivity model in attempts to quantify ligand 
effects without success. Earlier studies on octahedral l93 lr(lll) 
compounds 8 resulted in p.c.s. and p.q.s. values of little predictive 
use. For example, bis phosphine compounds had significantly higher 
C.S. values when the phosphines were cis than when trans. The spectra 
of Au(l) complexes were used to distinguish geometries; H ' r * 9 however, 
the p.c.s. and p.q.s. values of ligands calculated for the linear 
structures could not predict the spectra of other linear compounds 
not in the original correlation. The ligand orderings of p.q.s. 
obtained from LAuCl were different than those obtained from [AuL 2 ] . 
rr-bonding was shown not to be important for Au(l) complexes, so 
lack of predictability results from breakdown of the basic assumption 
of the additivity model. 10 It is believed that the gold atom con- 
centrates more 6s character into the bonds with strong donors. It 
is also claimed that a particular ligand' s donor strength depends 
on the other ligands bonded to the central atom. 

The effects described above have been offered as an explanation 
of the trans influence 11 and in compounds of higher coordination 
(e.g., Ir(IIl) octahedral complexes) an additional cis influence 
exists. The trans influence, which is an ordering of the ability 
of a ligand to weaken the bond of the group opposite it in the 
equilibrium state of a molecule, 12 has been studied by many physical 
methods. 13 The ligand orders vary according to the method used, 
partly due to the limitations of using bond lengths, stretching 
frequencies or NMR coupling constants as an indication of bond 
strength. A major problem arises in that most of the complexes used 
to study the trans influence contain cis ligands which introduces 
a cis influence, whose effect in both direction and magnitude on 
the bond studied is not well understood. 11 * 14 ' 15 



31 

A recent study 16 of this effect using 35 C1 nuclear quadrupole 
resonance (NQR) spectroscopy on compounds of the formulation LAuCl 
has the advantage of the absence of the cis influence. The 35 C1 
resonance is very sensitive to the trans ligand, L, and the NQR 
results could be correlated with the~~ THT "Au Mossbauer data. The 
temperature dependence of the NQR, which reflects tt bonding, 1 " 7 ' 18 
indicated increasing Cl~ tt donation to gold as the L group became 
a better tt acceptor. The trans influence order, derived from the 
35 C1 resonant frequency 19 extrapolated to 0°K was: PEt 3 , P(C 6 H 1:L ) 3 
> P(OMe) 3 > PPh 3 > P(OPh) 3 > AsPh 3 > 2-MeOC 6 H 4 NC > PhNC > SMe 2 > 
C 8 HnNC > PC1 3 > PPh 3 S > py > CI , 16 

The combined l97 Au Mossbauer and 35 C1 NQR results confirm the 
existence of delocalized bonding effects in linear Au(l) compounds. 
The importance of n bonding in these compounds is not resolved as 
the Mossbauer and NQR data do not give similar results. The con- 
clusions do indicate that the additivity model can not be applied 
to systems where the trans influence is an important consideration. 

References 



1. R. L. MBssbauer, Naturwissen schaften, 4j5, 538 (1958). 

2. G. M. Bancroft and R. H. Piatt, Adv. Inorg. Chem. Radiochem., 
15, 59 (1972). 

3. G. M. Bancroft, Coordination Chem. Rev., 11, 247 (1973). 

4. A. P. Marks, R. S. Drago, R. H. Herber and M. J. Potasek, 
Inorg. Chem., 15, 259 (1976). 

5. G. M. Bancroft and E. T. Libbey, J. Chem. Soc. Dalton, 
2103 (1973). 

6. C. A. McAuliffe, R. U. Parish and P. D. Randall, 
J. Chem. Soc. Dalton, 1426 (1977). 

7. P. G. Jones, A. G. Maddock, M. J. Mays, M. M. Mulr and 
A. F. Williams, J. Chem. Soc. Dalton, 1*134 (1977). 

8. A. F. Williams, G. C. H. Jones and A. G. Maddock, 
J. Chem. Soc. Dalton, 1952 (1975). 

9. G. C. H. Jones, P. G. Jones, A. G. Maddock, M. J. Mays, 
P. A. Vergano and A. F. Williams, J. Chem. Soc. Dalton, 
1440 (1977). 

10. M. G. Clark, A. G. Maddock and R. H. Piatt, J. Chem. Soc. 
Dalton, 281 (1972). 

11. S. S. Zumdahl and R. S. Drago, J. Amer. Chem. Soc, 90, 
6669 (1968). 

12. A. Pidcock, R. E. Richards and L. M. Venanzi, 
J. Chem. Soc. A, 1707 (1966). 

13. T. G. Appleton, H. C. Clark, and L. E. Manzer, 
Coordination Chem. Rev., 10, 335 (1973). 

14. B. W. Dale, R. J. Dickenson and R. V. Parish, 
J. Chem. Soc. Chem. Comm. , 35 (1974). 

15. P. B. Hitchcock, B. Jacobson and A. Pidcock, 
J. Chem. Soc. Dalton, 2044 (1977). 

16. P. G. Jones and A. F. Williams, J. Chem. Soc. Dalton, 
1430 (1977). 

17. T. E. Haas and E. P. Marram, J. Chem. Phys . , 43, 3985 (1965). 

18. T. L. Brown and L. G. Kent, J. Phys. Chem., JWJ 3572 (1970). 

19. C H. Townes and B. P. Dailey, J. Chem. Phys., 17, 782 (1949). 

General Reference 

R. S. Drago, Physical Methods in Chemistry ; Chapts. 14 and 15, 
Saunders Company, Philadelphia, PS ( l^Y Y ) . 



32 

POLYMER BOUTO SCHIFF BASE COMPLEXES 

John Gaul January 26, 1978 

Two types of base containing polymers were prepared to investi- 
gate the question of site separation 1 > 2 > 3 in coordinate polymer com- 
plexes. Co(ll) and Cu(ll), four coordinate complexes, were incor- 
porated into pyridine and imidazole containing resins through axial 
coordination to the base residues. The ability of the pyridine 
polymer to coordinate the Co(ll) and Cu(ll) species was examined and 
found to be limited whereas the imidazole polymer binds these species 
very strongly. Co(DPGB) 2 # 2MeOH was incorporated into the imidazole 
polymer and the extent of coordination by the base residues examined 
by ESR. This Co(ll) complex is axial and contains one electron in 
the d 2 orbital. Axial coordination of one or two nitrogen bases 
can bi distinguished by ESR. 4 Incorporation of this cobalt complex 
yielded characteristic 1:1 and 2:1 imidazole-cobalt adducts depending 
on the loading of the imidazole. The implication for these types of 
polymers and complexes is that the loading is more important in 
achieving site separation than is crosslinking. 

In addition to the monodentate chelate polymers, the synthesis 
and characterization of novel multidentate chelating polymers was 
attempted. Polymer bound complexes based on the H 2 SalDFT, H 2 SalDAP, 
H 2 SalDAPP and PicDPT ligand systems were prepared through chelation 
of Mn(ll), Fe(lll), Fe(ll), Co(ll), Ni(ll), Cu(ll) and Zn(ll). The 
ability of the cobalt complexes to reversibly bind 2 was verified 
by ESR. The analogous parent complexes were prepared by published 
procedures 5 * 6 as an aid in identification. The heretofore unreported 
complex Fe(SalMeDPT) was also prepared and characterized by 
Mttssbauer analysis. 

The polymer complex (P)-Co(SalDPT) was used in catalytic oxi- 
dations of substituted phenols. Both macroreticular and micro- 
reticular resin catalysts were employed. In general, the oxidations 
with the polymer catalyst were slower than those with the parent 
complex. However, a great enhancement in product selectivity over 
the parent complex could be achieved with only small changes in the 
resin loading of the catalyst. The possible effects the polymer 
is exerting on the catalytic action of the polymer bound complex, 
to cause such product selectivities, is discussed in terms of what 
is known about phenol oxidations by cobalt oxygen carriers. 7 "' 8 

References 

1. J. I. Crowley, T. B. Harvey, III, H. Rapoport, J. Macromol. 

Sci. Chem., A7(5), 1117 (1973). 

2. J. P. Collman, C. A. Reed, J. Am. Chem. Soc, 95, 2048 (1973). 

3. L. T. Scott, J. Rebek, et. al. , ibid . , 99 , 625^1977). 

4. B. S. Tovrog, Ph.D. Thesis, University of Illinois (1975). 

5. L. Saccooni, I. Bertini, J. Am. Chem. Soc, 88, 5180 (1966). 

6. R. D. Patton, L. T. Taylor, Inorg. Chim. Acta, 8, 191 (197*0 . 

7. A. Nashinaga, K. Watanabe, T. Matsuura, Tetrahedron Letters, 14 , 

1291 (1974). 

8. R. A. Sheldon, J. A. Kochi, Adv. Catalysis, 274 (1976). 



33 

TEMPERATURE PROGRAMMED DESORPTION AS A TOOL IN CATALYSIS 

Steven J. Hardwick February 16, 1978 

Although reactions occurring in solution are routinely investigated 
by chemists, it is often difficult to get detailed information 
about reactions which occur on surfaces. Because of the importance 
of heterogeneous catalysts in industrial processes, however, there 
is a great deal of interest in heterogeneous surface chemistry. 
The first step toward understanding these reactions is to learn 
how reactants interact with the surface. A relatively simple 
technique for studying the interaction of molecules with a surface 
is temperature programmed desorption, (TPD). 

In TPD, a gaseous reactant is allowed to adsorb on the 
surface of a catalyst. Physically adsorbed gas is removed and 
the rate of desorption of chemisorbed gas is mea&ured as a 
function of temperature. The desorbed gas is typically measured 
by a thermal conductivity detector. -^"^ Alternatively, the desorbed 
gas may be monitored by a mass spectrometer . 4 > 5 The output of 
a TPD experiment is called a desorption chromatogram, 1 which is 
a plot of recorder response versus temperature. The recorder 
response is proportional to the concentration of desorbed gas 
and the rate of desorption. Desorption chromatograms can provide 
information about the number of sites which chemisorb a given 
reactant, and the relative population of these sites. They can also 
give information about the strength of surface-adsorbate interactions. 
Finally, under favorable conditions they can provide information 
about which chemisorption sites are catalytically active. 

The type of information which is available from TPD, can 
be illustrated by results obtained from the ethylene-alumina 
system. TPD of ethylene chemisorbed on y-alumina yields two 
peaks corresponding to two active chemisorption sites. 6 5 7 it is 
found that reaction with hydrogen gas at temperatures below 
60°C leads eventually to the hydrogenat ion of all the ethylene 
adsorbed on site I. This indicates that site I, which comprises 
less than 1.6% of the surface of alumina is responsible for the low 
temperature hydrogenation of ethylene. 

In order to put the interpretation of desorption chromatograms 
on a more quantitative level, one would like a means of relating 
T m , the temperature corresponding to a peak maximum, to the strength 
of a surface-adsorbate interaction. For the simplest case, where 
there is first order desorption from an energetically homogeneous 
surface, and where no readsorption takes place, the following 
functional equation is used to obtain the activation energy of 
desorpt ion . *■ 



34 

21og T m - log 3 = E d /2.303RT m + log(E d /AR) 

In this equation 3 is the heating rate, E d is the activation energy 
of desorption and A is the Arrhenius constant. If one plots 
21og T m - log vs. 1/Tjn for a wide range of 3 a straight line 
is obtained, from which Ed and A can be determined. Interpretation 
of TPD chromatograms can be complicated by readsorption , 1 diffusion 
from porous surfaces,! heterogeneous surfaces,^ and second order 
desorption . ^ ?10 Theoretical treatment of these problems have been 
made, and are included in the above references. 

A detailed example of the use of TPD is given by a series 
of Pt catalysts. Pt is an active hydrogenation catalyst. For 
this reason a knowledge of how ethylene and hydrogen interact 
with platinum is desirable. Four forms of chemisorbed hydrogen 
are obtained from TPD of hydrogen adsorbed on Pt black. H A 
fifth hydrogen peak has been observed by other workers, however 
this data may be suspect. 12 

Ethylene is irreversibly adsorbed on Pt/silica and Pt black. 
TPD leads only to self-hydrogenation and decomposition. 13 ,5 
The activation energies of self-hydrogenation and decomposition 
are 11.1 and 37.3 Kcal/mole respectively. If hydrogen is 
chemisorbed on Pt , and allowed to react with ethylene, it is 
found that only the y , and possibly the 3 forms of chemisorbed 
hydrogen participate in the hydrogenation of ethylene. 1^ 

It is interesting that by alloying Pt with Sn one obtains 
two ethylene desorption peaks, and that the self-hydrogenation and 
decomposition reactions do not occur. ° This is attributed to a 
dilution of Pt atoms by Sn . This dilution is called an "ensemble 
effect. "15 The intensity of the high temperature peak decreases 
at increasing Sn concentrations. This is also attributed to the 
"ensemble effect." In addition, increasing the concentration of 
Sn leads to a decrease in the temperature corresponding to the peak 
maximum. This is attributed to a "ligand effect. "15 

In summary, TPD is a simple technique which can provide 
a great deal of information about practical catalysts. New and 
improved theoretical treatments as well as improvements in the 
apparatus itself should lead to greater applicability to the study 
of metal catalysts. Furthermore, coupling of TPD with TGA, DTA , 
IR, and other techniques should lead to a better understanding 
of reactions on the surface of catalysts. 



35 



References 



1. R. J. Cvetanovich and Y. Amenomiya, Advan. Catal ., 17, 103 (1967). 

2. R. J. Cvetanovich and Y. Amenomiya, Catal. Rev., 6^ 21 (1972). 

3. R. J. Cvetanovich, Amer. Chem. Div. Petrol. Chem. (prepr.) 
17,C77 (1972). 

4. A. W. Smith, J. M. Quets, J. of Catal., 4, 163 (1965). 

5. H. Verbeek, W. M. H. Sachtler, J. of Catal., 4_2, 257 (1976). 

6. Y. Amenomiya, R. J. Cvetanovich, J. Phys . Chem., 6T7, 144 (1963). 

7. Y. Amenomiya, J. H. B. Chenier, R. J. Cvetanovich, J. Of Catal., _9, 
28 (1967). 

8. G. Carter, Vacuum, 12, 245 (1962). 

9. J. Konvalinka, J. J. F. Scholten, J. C. Rasser, J. of Catal., 48 , 
365 (1977). 

10. F. M. Lord, J. S. Kittelberger , Surface Sci., 4_3_, 173 (1974). 

11. S. Tsuchiya, Y. Amenomiya, R. J. Cvetanovich, J. of Catal., 19 , 
245 (1970). 

12. E. Moger, M. Hegedus , G. Bensenyei, F. Nagy, Reac . Kinet . and 
Catal. Lett., _5, 73 (1976). 

13. R. Komers , Y. Amenomiya, R. J. Cvetanovich, J. of Catal., 
15 , 293 (1969). 

14. S. Tsuchiya, M. Nakamura , J. of Catal., 50_, 1 (1977). 

15. W. M. H. Sachtler, Le Vide 164, 67 (1973). 



36 

THE MIXED VALENCE CHEMISTRY OF SOME DISCRETE POLYNUCLEAR 

IRON AND RUTHENIUM COMPOUNDS 

Susan L. Lambert February 21, 1978 

INTRODUCTION 

Mixed valence chemistry is concerned with compounds containing 
two different formal oxidation states of the same element. Many 
naturally occurring minerals are mixed valence compounds. Because 
of their unique characteristics, mixed valence compounds attracted 
early attention. Mixed valence compounds generally exhibit an in- 
tense visible/nir transition arising from the intervalence transfer 
of an electron. It is this absorption band which gives many mixed 
valence compounds their characteristic dark coloration. This ab- 
sorption is attributed to the transition (2,3) "(3,2)*, i.e., the 
Frank-Condon barrier to electron transfer required for a transition 
between two centers with the same chemical composition but differing 
in environment due to their different charges. 

By 1922, the basic phenomenon of valence oscillation in mixed 
valence species had been elucidated. 1 Mixed valence chemistry perme- 
ates many areas of current interest and many relatively new physical 
techniques, such as MBssbauer, are being used to study mixed valence 
enzymes, mixed valence reaction intermediates, and electron transfer 
in mixed valence compounds. 

In 1967, Robin and Day 2 set forth a classification scheme for 
mixed valence compounds on the basis of ground state delocaiization 
of the odd electron. Day's scheme contains three classes of mixed 
valence compounds: those which have no delocaiization in the ground 
state, class lj those with some delocaiization in the ground state 
or class 2; and, class 3 compounds which experience maximal delocaii- 
zation of the odd electron in the ground state. Thus, electron ex- 
change between two atoms of different valences can vary from very 
fast to extremely slow. The properties of these compounds have been 
found to vary with the extent of delocaiization. Class 1 compounds 
have two atoms in different valence states which do not interact. 
No intervalence transfer (IT) band is seen. Class 3 compounds, be- 
cause of the delocaiization of the electron, exhibit transitions 
from a delocalized bonding molecular orbital to an antibonding molec- 
ular orbital in the excited state. There is no thermal barrier to 
electron transfer in the ground state of a class 3 compound. 

N. S. Hush has advanced a theory 3 to explain the properties of 
class 2 compounds. The energy of the IT band is used to calculate 
the rate of thermal electron transfer and the IT band intensity is 
used to evaluate the extent of ground state delocaiization. 



o J 



RUTHENIUM COMPOUNDS 

Many physical techniques have been applied to mixed valence 
ruthenium dimers to study their electronic exchange. In the case 
of bis(pentaammineruthenium) pyrazine, 4 such varied techniques as 
optical spectroscopy, 4 Mossbauer, 5 NMR, 6 EPR, 6 IR, 7 and X-ray 
crystallography 8 have been used to determine the rate at which the 



37 

electron is exchanging. The application of a battery of physical 
techniques is common in characterizing a mixed valence system. In 
determining the rate of thermal electron transfer, the inherent 
time resolution of the different methods is used. In a series of 
bis(pentaammineruthenium)L compounds where L is variously pyrazine, 
4,4' bipyridyl and cyanogen, Taube and coworkers 4 * 3 > c reported on 
the extent of derealization based on optical spectroscopy where the 
time resolution is 10 13 sec. Only the cyanogen bridged species is 
delocalized on this time scale. In this compound only one CN stretch 
is seen at a frequency intermediate between that for the Ru 3 and the 
Ru 2 compounds. This indicates that in 10'" 13 sec the cyanogen bridge 
cannot distinguish between the two ruthenium centers. In the pyra- 
zine and 4,4' bipyridyl bridged compounds,, the electron is exchanging 
slow enough so that bri-dging ir frequencies show the presence of 
two distinct ruthenium sites. 

IRON COMPOUNDS 

Biferrocene derivatives have proved particularly amenable to 
mixed valence studies. Biferrocene, biferrocenylene and [1.1] ferro- 
cenophane have all been prepared as mixed valence species. These 
compounds have been studied by polarography, 9 visible/ir spectro- 
scopy, 10 M5ssbauer, 11 EPR.. - ll and magnetic susceptibility. 11 Related 
series of these compounds have been studied and correlations with 
Hush's theory have been made. 12 The IT band maxima of unsymmetrically 
substituted biferrocenes are blue shifted from the maxima of sym- 
metrically substituted biferrocenes. This is expected since in un- 
symmetrically substituted species electron exchange resulting in an 
unfavorable valence isomer will require more energy. 

CONCLUSIONS 

There are many mixed valence systems yet to be investigated. 
The iron-sulfur proteins contain two valence states of iron in poly- 
nuclear clusters. Model compounds are being investigated, but the 
proteins themselves are not as easily studied. Elucidation of the 
mechanism of electron exchange in these species will enable us to 
better understand electron transport in biological systems. Also, 
many mixed valence compounds are being looked at today as possible 
new semiconductors and even high temperature superconductors because 
of the enhanced conductivity of many mixed valence compounds. 

REFERENCES 

1. H. L. Wells. Am. J. Sci. , 3, 4 17 (1922). 

2. M. Robin, P. Day, Adv. Inorg . Chem. Radiochem . , _10> 247 (1967). 

3. N. S. Hush, Prog . I no r gT ~ CHem . 7~^7 391 (l°b>>). 

4. C. Creutz, H. Taube, J. Am. Chem. Soc., 91, 3n^8 (196^). 
4a. ibid., 95, 10R6 (19737. 

4b. ibid., 9o, 7< Q '27 (1974). 

5. C. Creutz, M. L. Good, S. Chandra, Inorg . Nucl. Chem . Lett . , 

9, 171 (1Q73). 

6. B. C. Bunker, R. S. Drago, D. N. Hendrickson, S. L. Kessel, 

R. Richman, J. Am. Chem. Soc, in press. 

7. J. K. Beattie, N. S. Hush, P. Taylor, Inorg . Chem . , 15, Q92 (l°76). 

8. J. K. Beattie, JCS Dalton, 1121 (1977). 



3S 



9. W. H. Morrison, S. Krogsrud, D. N. Hendrickson, Inorg . Chem . , 12 , 
1998 (1973). 

10. D. 0. Dowaine, C. LeVande, J. Park, F. Kaufman, 
Acct . Chem . Res. , 6, 1 (1973). 

11. W. H. Morrison, D. N. Hendrickson, Inorg . Chem . , lh , 2331 (1975). 

12. C. LeVande, K. Bechgaard, D. 0. Cowain, M. D. Rausch, 
J. Am. Chem. Soc. , oq, 2964 (1977). 






39 

"METAL-QUINONE COMPLEXES" 

Michael W. Lynch March 2, 1978 

Introduction 

Metal-quinone interactions have been a topic of considerable 
interest. Quinones are postulated to play important roles in many 
biological systems. For example,, ubiquinones are an integral part 
of the respiratory electron transport chain, and plasto qui nones are 
involved in photosynthesis^ 1 Quinones can act as either one or two 
electron acceptors. (Q f SQ T % HQ~ 2 . ) Due to this interesting 
redox chemistry, quinones are thought to be "unique" ligands for 
inorganic complexes. The following are the three general types of 
metal-quinone interactions: TT-tyne with para-quinones, q-type with 
either para-quinones or ortho-quinones and finally donor-acceptor 
complexes which will not be discussed in this seminar. 

Physical Properties 

Metal-quinone compounds can be characterized by the following 
physical techniques: infrared (CO stretch), EPR, NMR, magnetic 
susceptibility, M5ssbauer, electrochemistry and X-ray crystallography, 
The majority of the inorganic complexes have been identified by the 
characteristic shift to lower energies of the infrared v band re- 
sulting from coordination to a metal. The carbon-oxygen stretching 
vibrations of ortho-quinones shift 250 cm 1 to 300 cm 1 . p-Quinones 
experience shifts of 20 cm" 1 to 100 cm x with n-bonding, and shifts 
of 180 cm" 1 to 250 cm" 1 with a o-type interaction. 2 

TT-Complexes 

The n-complexes between para-quinones and metals are formed 
when the quinones act as "diolefins." The metal bonds with the rr 
electron density of the "ring" rather than direct interaction with 
the oxygen atoms. The compound (rr-duroquinone) iron tricarbon^l is 
an example of a complex in which the quinone is acting as a le (r| 4 ) 
donor to the iron. 3 ^ Bodner has prepared a r) 6 -hydroquinone complex 
[(Ci O H 12 02)m(C 5 H5) ] +2 in solution and characterized it by NMR. The 
data indicate reduction of the 1,4-benzoquinone moiety has occurred, 
allowing it to act as a 6e~ donor. 3) ° There are no reports in the 
literature of rr-complexes of ortho-quinones. 

o-Complexes 

Both ortho-quinones and para-quinones bond to metals in a a- 
fashion ( i.e ., through the oxygen atoms). Numerous o-complexes of 
para-quinones have been reported. One of the first to be reported 
was K 6 [(NC) 5 Co(c 6 H40 2 )Co(cn) 5 ]. In this complex the dianion form 
of 1, 4-benzohydroquinone is postulated to be bridging between the 
two pentacyano cobaltate(lll) groups. 4 Calderazzo and coworkers 
have prepared some Schiff base complexes of cobalt and iron. They 
are [Co( salen) (py) ] 2 C 6 H 4 0o and [Fe( salen) ] 2 -quinone where salen = 
N, N' -ethylenebis( salicylideneirainato) . 5 > 6 These compounds have been 



40 



extensively studied with IR and magnetism. Interesting magnetic 
properties arise for the cobalt complex [ (u £ ~ '^ 0.74 B.M. at 287K) 
and the iron complex (p ff -5.76 B.M. at 297 K) . These complexes 
consist of the dianion of the hydroquinone bridging between two 
metal(lll) centers. In the case of the iron compounds weak anti- 
ferromagnetic exchange interactions (-6.1 > J > -0.19) are indi- 
cated by low temperature susceptibility results. 

There are indeed many examples of orthoquinone-metal complexes 
found in the /literature. One finds ortho-quinones coordinated to 
metals in the three possible forms: neutral quinones, semi-quinones, 
and hydroquinone dianions. The catechol complex (3, 5-di-tert-butyl- 
catecholato) (triethylenetetramine) cobalt (ill) chloride is an example 
of the hydroquinone dianion form." 7 This complex is characterized by 
two prominent infrared bands (the C-C skeletal vibration at 1480 cm -1 
and the carbon-oxygen stretching vibration at 1250 cm" 1 ). Electro- 
chemical studies on the compound indicated a stable "one electron" 
oxidation product. The isolated species, [Co(trien) (DB ) ]C1 2 , was 
thus formulated with a coordinated semiquinone . The presence of a 
semiquinone is supported by infrared and EPR data. The carbon-oxygen 
stretching vibration is at 1440 cm" 1 . Solution and solid EPR spectra 
show an eight line hyperfine pattern attributed to 59 Co (i = 7/2). 
The cobalt hyperfine coupling constant (A r = 9.67 gauss) is rather 
small in comparison to other Co(ll) completes. This indicates the 
unpaired electron resides in a molecular orbital localized mainly on 
the semiquinone ligand. 



8 



An example of a neutral quinone complex is found in the work of 
Floriani. 9 [TiCl 4 (quinone) ] was prepared specifically to study the 
perturbation of a metal interacting with a neutral quinone species 
(in this case quinone = 9,10 phenanthrenequinone) . No shift in the 
carbon-oxygen stretching vibration was observed in the infrared. 9 

Biological Systems 

Ubiquinones are believed to be an important link in the res- 
piratory electron transport chain, playing an essential role in the 
enzymatic reactions of succinate dehydrogenase. EPR results indicate 
an interaction between an ubisemiquinone and a second paramagnetic 
species. Three candidates for this second species have been hypo- 
thesized: a second ubisemiquinone, a f lavin-semiquinone, and a 
Fe-S center (namely S 3 ). Ruzika and coworkers have concluded from 
their studies that the second species is a f lavin- semiquinone. 10 
Ohnishi, on the other hand, supports the hypothesis of a Fe-S center 
(S3). 11 These systems have been studied extensively and are still 
not completely understood. 

Conclusions 



The vast area of metal-quinone interactions have been studied 
for some time, but only the beginnings of this fruitful area have 
been harvested. This should be an open invitation to inorganic 
chemists to prepare and characterize inorganic model complexes as 
an aid to further the understand ing of metal-quinone interactions 
in biological systems. 



41 



References 

1. T. P. Singered, "Biological Oxidations," Interscience, pg. 533 

(1968). 

2. Patai, "The Chemistry of Quinonoid Compounds," Interscience, 

pg. 257 (1974). 

3a. Sternberg, Markby and Wender, Jour . Am. Chem. Soc. , 80, 1009 
(1958)., — 

3b. Bodner, Jour. Organometal . Chem . , 88 , 391 (1975). 

4. Vlcek and Hanzlik, Inorg . Chem . , 6, 2053 (lQo7). 

5. Floriani, Fachinetti and Calderazzo, JCS Dalton , 765 (1973). 

6. S. L. Kessel and D. N. Hendrickson, submitted for publication. 

7. D. Brown and P. A. Wicklund, Inorg . Chem . , 15, 396 (1976). 

8. D. Brown, P. Wicklund and L. Beckmann, Inorg . Chem . , 15, 1996 

(1976). 

9. Floriani, Fachinetti and Calderazzo, JCS Dalton , 765 (1973). 

10. F. Ruzica, H. Beinert, K. Schepler, R. Dunham and R. Sands, 

Proc. Nat. Acad . Sci . USA , 72, 2886 (1975). 

11. T. Ohnishi and W. Ingledew, F.E.B.S. Letters, 54, 167 (1975). 



42 

ELECTROCHEMICAL REDUCTION OF ? ON PHTHALOCYANINE 

COMPLEXES 

Charlotte Owens March 7, 1978 

Because of the need to develop the technology for new energy 
sources and to improve that of present ones , there is increased 
interest in fuel cells. As direct converters of chemical to 
electrical energy, fuel cells are not subject to Carnot cycle 
limitations. ' Generally, the oxidant in fuel cells is 02 , which is 
attractive because of its availability in air; however, the oxygen 
electrode is also the major problem in fuel cells because the 
sluggishness of the electrode reaction O2 + 4e~ ■*■ 2H2O lowers the 
cell voltage and the power output. l'^ 

The most widely used electrode materials have been noble metals 
such as platinum; however, even platinum, one of the best electro- 
catalysts available for O2 reduction, performs poorly in comparison 
to its ability to catalyze other electrochemical reactions. Because 
of the cost and low world supply of platinum, wide-spread use of 
fuel cells requires the development of other electrode materials, 
especially non-metallic ones. 2 Some of the other materials which 
have been tested are various oxides of the spinel and perovskite 
crystal types, tungsten bronzes, disulfides, thiospinels , and 
various metal chelates, including tetraphenyl porphyrin and 
phthalocyanine complexes! 3 Some of the studies of the activity 
of phthalocyanine complexes will be examined here. 

Various metal chelates of N14 (tetraaryl porphyrins, dibenzo- 
tetraazaannulenes , phthalocyanines ) , N2O2 (polymeric "saloph" 
complexes), and N2S2 [ diacetyldi(thiophenylhydrazones ) ] donor sites 
have been investigated as electrode materials; only the complexes 
of N14 donor sites, all of which are macrocyclic and have extensive 
ir-conjugation , were electrochemically active for the reduction of 
O2 • The electrochemical process evidently is not the reduction of 
a transition metal complex-02 adduct formed in solution, as the 
"saloph" complexes, known to reversibly bind O2 , were inactive; 14 
also, the diffusion limiting current for reduction of O2 on 
Co ( II )tetrasulfonatephthalocyanine adsorbed on graphite was found 
to be independent of the concentration of complex in solution. 
As electrode materials, the most active phthalocyanine complexes 
are polymeric, and contain iron or cobalt ;6 their activity, which 
depends greatly on the support, is best on high surface-area carbon. ^ 
The activity of these complexes has been tested in acidic, basic, 
and neutral solutions; while several electrodes containing iron 
phthalocyanines were found to be unstable in acid, 14 Meier and coworkers 
have made electrodes which are stable, at a constant current flow 
of 20mA/cm2, for 3000 hours . 8 Through use of a radioisotopic tracer, 
Meier monitored the dissolution of iron into 6N H2SO4 solution, and 



found no direct correlation with the potential drop of the cell with 

time. 8 Various correlations between electrocatalytic activity and 

electrochemical oxidation potentials, magnetic properties, 

• • 9 — 1 9 • « ti to 

catalytic activity, and conductivity 11 5 1Z have been offered. 

p "i ^ — n r 
In attempts to learn about the active sites, Mossbauer, ' 

optical spectroscopy , 13 and ESCA^ have been used. Some of the 

studies have paid insufficient attention to the bulk vs. molecular 

properties, and the vast effect of ambient conditions (C^i^O) 

on the conductivity and behavior of the phthalocyanine complexes. 

(References discussing these effects include 16 and 17). 

General References 

1. H. A. Liebhafsky, E. J. Cairns, "Fuel Cells and Fuel Batteries," 
John Wiley £ Sons, New York, 1968. 

2. J. O'M. Bockris, S. Srinivasan, "Fuel Cells: Their Electrochemistry," 
McGraw-Hill, New York, 1969. 

Literature References 

3. Ernest Yeager, Nat ' 1 . Bur . Stand. Spec . Publ . 4 5 5, 203 (1976). 

4. H. Alt, H. Binder, G. Sandstede, J. Catal . 28 , 8~Tl973). 

5. Jose Zagal, Rajat K. Sen, Ernest Yeager, J. Electroanal . Chem . 
83_, 207 (1977). 

6. A. J. Appleby, M. Savy, Electrochim . Acta 21, 567 (1976). 

7. R. J. Brodd, V. Z. Leger, R. F. Scarr, A. Kozawa, Nat ' 1. Bur . Stand . 
Spec . Publ . 455 , 253 C1976). 

8. H. Meier, U. Tschirwitz, E. Zimmerhackl, W. Albrecht , G. Zeitler^, 
J. Phys . Chem . 81, 712 (1977). 

9. J. -P. Randin, Electrochim . Acta 19 , 83 (1974). 

10. J. Manassen, J. Catal . 3_3, 133 (1974). 

11. H. Meier, W. Albrecht, U. Tschirwitz, E. Zimmerhackl, 
Ber . Bunsenges . Phys . Chem . 77 , 843 (1973). 

12. H. Meier, U. Tschirwitz, E. Zimmerhackl, W. Albrecht, 
BMVg-FBWT-75-6 (1975). 

13. A. J. Appleby, J. Fleisch, M. Savy, J. Catal . 44, 281(1976). 

14. A. J. Appleby, M. Savy, paper 345, 151st Meeting of the 
Electrochem. Soc. Philadelphia, PA, May 1977. 

15. Ragnar Larsson, Jiri Mrha , Jan Blomqvist, Acta Chem. Scand. 26, 
3386 (1972). — ' 

16. Hiroyasu Tachikawa, Larry R. Faulkner, J. Am. Chem . Soc . 100 
(in press). 

17. S. E. Harrison, K. H. Ludewig, J. Chem. Phys. 45, 343 (1966). 






44 



SYNTHESIS AND STEREODYNAMICS OF CYC LO PENT ADIENYL RHODIUM 

AND IRIDIUM CLUSTER COMPLEXES 

R. Joe Lawson March 9, 1978 

Recently it was reported 1 ' 2 that trimethylamine oxide (Me 3 NO) 
would react with coordinated carbonyls of transition metal complexes 
producing C0 2 and a coordinately unsaturated complex. This provides 
a method for inducing relatively quantitative carbonyl loss under 
mild conditions. Since CO loss in the absence of additional ligands 
may lead to metal-metal bond formation, 3 we have investigated the 
use of Me 3 NO to facilitate cluster formation from coordinately 
saturated monometallic carbonyl complexes. 

The reaction of Me 3 NO and CpRh(CO) 2 (Cp=r] 5 -cyclopentadienyl) 
produces the known complexes Cp 2 Rh 2 (CO) 3 , C3 V -Cp 3 Rh 3 (C0)3 and 
Cs-Cp 3 Rh 3 (CO) 3 in much better yields than were obtained by photolysis 
of CpRh(CO) 2 . 4 ' s ' 6 In addition, a previously unknown species, 
Rh 4 Cp 4 (C0) 2j was isolated. 

Rhodium- rhodium coupling constants 7 have been derived from the 
analysis of high resolution proton NMR spectra of Cp 2 Rh 2 (CO) 3 , 
Cp 2 Rh 2 (NO) 2 and Cp 2 Rh 2 (uCH 2 ) (CO) 2 . The magnitude found for 1 J Rh _ Rh ^ 
4-5 Hz, is consistent with that which would be predicted for a 
simple sigma bond between two rhodium atoms and a coupling mechanism 
dominated by the Fermi contact term. 

The improved syntheses of the C 3V and Cs isomers of Cp 3 Rh 3 (CO) 3 
have made it practical to obtain 13 C0 enriched samples which were 
necessary for solution characterization. Both isomers are fluxional 
in solution without isomerization. A novel carbonyl scrambling 
mechanism has been proposed to explain this behavior. 8 j 9 The car- 
bonyls of Cp 4 Rh 4 (C0) 2 are also fluxional at ambient temperatures. 

Reacting Me 3 NO with CpIr(CO) 2 yields Cp 2 Ir 2 (C0) 3 , Cp 3 Ir 3 (CO) 3 
and Cp4lr 4 (C0) 3 . The total yield of isolated species, however, is 
less than with CpRh(C0) 2 . 

References 

1. Y. Shro and E. Hazum, J. Chem. Soc. , Chem. Commun. , 829 (1975). 

2. V. Koell, J. Organometal. Chern . , 133 , 53 (1977). 

3. K. P. C. Vollhardt, J. E. Bercaw and R. G. Bergman, 

J. Organometal. Chem ., 97 , 283 (1975). 

4. 0. S. Mills and E. F. Paulus, J. Organometal. Chem ., 10 , 

331 (1967). 

5. E. F. Paulus, E. 0. Fischer, H. P. Fritz and H. Schuster-Woldan, 

J. Organometal. Chem ., 10 , P3 (1967). 

6. J. Evans, B. F. G. Johnson, J. Lewis and J. R. Norton, 

J. Chem. Soc, Chem. Commun ., 79 (1973). 

7. R. J. Lav/son and J. R. Shapley, submitted for publication. 

8. R. J. Lawson and J. R. Shapley, J. Amer. Chem. Soc , 98 , 

7433 (197b). 

9. R. J. Lawson and J. R. Shapley, Inorg. Chem ., 17 , 0000 (1978). 



^5 

POLYMER SUPPORTED HOMOGENEOUS CATALYSTS 

David Corbin March 14, 1978 

Catalysts are important in many industrial processes, such as 
petroleum refining, petrochemical manufacture, coal and coal tar 
derivative processing, and fertilizer manufacture. 1 Generally, 
catalysts are divided into two categories: heterogeneous and homo- 
geneous. Almost all industrial catalysts are heterogeneous in 
nature, that is, the reaction takes place at a phase boundary. 
During the past several years, homogeneous catalysts have received 
a great deal of attention. These are catalysts which are soluble 
in the liquid phase reactant. Homogeneous catalysts have received 
limited industrial use — chiefly because of the difficulty of sepa- 
ration from the reaction products. 

Recently, an intermediate system has offered a great deal of 
promise. Here, the active homogeneous catalyst is firmly held on 
an insoluble, porous polymeric support. Through such an attachment, 
it was hoped that the advantages of the homogeneous catalyst would 
be retained and the disadvantages removed. Many such systems have 
been studied. 2 ~ 6 

One type of polymeric support that is commonly used is that of 
polystyrene-divinylbenzene copolymer resins. By varying the amount 
of divinylbenzene and hence the crosslinking and pore size in the 
polymer, one changes the activity and the selectivity of the attached 

catalyst. 

Various functionalized polystyrene copolymers have been employed, 
Metal complexes have been attached to crosslinked polystyrene by 
amine, 7 phosphine, cyclopentadienyl, bipyridyl, 8 bound imidazole, 9 
and various Schiff bases, 10 > 1:L among other ligands. The complex is 
then easily attached by equilibration of the polymeric ligand with 
a similar or weaker complexing ligand on the metal. 

A system of great interest is that of the supported Wilkinson's 
catalyst, (PPh 3 ) 3 RhCl. This complex catalyzes the reduction of 
olefins by hydrogen at conveniently measurable rates at atmospheric 
pressure and temperature. While the species attached to a 2.% cross- 
linked resin was found to be about O.Oo" times as active as the un- 
supported catalyst 12 (dependent upon the degree of crosslinking, 
bead diameter, temperature, and ratio of phosphine to rhodium), it 
was also shown to be selective toward smaller olefins. Overall ac- 
tivities in the hydrogenation of various types of olefins by the 
supported catalyst paralleled those reported for the conventional 
catalyst. 13 Moreover, the presence of olefin isomerization reactions 
and significant increases in the selectivity towards the reduction of 
sterically hindered olefins were observed for the supported 
catalyst. 



14 



Grubbs and coworkers 15 have described in detail the selectivity 
of the supported catalyst on the basis of the olefin size and the 
polar selectivity. Two opposing factors are considered. As the 
solvent polarity increases: l) the pore sizes decrease and 2) a 
polar gradient is established. 



46 

Variable selectivity of this supported catalyst was observed 
for primary and secondary olefins. This selectivity was found to 
be dependent on the mode of preparation and the presence of residual 
nonpolymeric ligands. Comparison was made of the ratio of reduction 
rate of 1-hexene/cyclohexene for these complexes and those of non- 
attached species. Using these comparisons, it was shown that the 
structure of the supported complex is a function of the loading of the 
metal on the polymer. 16 A more detailed structure of the attached 
species was obtained using EXAFS. 1Y 

Another system of interest is that of supported titanocene. 
Titanocene and its hydrides are useful catalysts in the hydrogenation 
of unsaturated hydrocarbons and molecular nitrogen. 18 j 19 j 20 One of 
the major problems with this catalyst is it readily dimerizes to form 
a catalytically inactive species which blocks the open coordination 
site required for homogeneous catalysis. To prevent this, Grubbs and 
coworkers 21 " 24 attached the titanocene precursor Cp 2 TiCl 2 j to a 20$ 
crosslinked polystyrene-divinylbenzene copolymer resin. Verification 
that only one species was present on the bead was made using elemental 
analysis and far infrared studies. The attached species was found to 
be about twenty times as active a hydrogenation catalyst as the corre- 
sponding non-attached species. Pulverized beads showed an increased 
activity of 8 to 10 times. This was due to the ability of the rigid 
polymeric matrix to keep the metal centers apart, thus preventing 
dimerization. This catalyst was also found to be more selective to- 
ward the smaller olefins. 

The supported titanocene system was tested for its ability to 
fix nitrogen. In the 20$ crosslinked resin, the centers were too 
far apart for the formation of the theoretical intermediate complex. 
By lowering the crosslinking, fixation was observed. 

Since it was proposed that the increase in activity for the 
supported titanocene resulted from site isolation on the polymer, 
site isolation was studied as a function of the loading of the 
catalyst. This study led to model for site isolation. 25 

Pittman and coworkers 26-28 have performed sequential multistep 
reactions catalyzed by polymer-anchored homogeneous catalysts. The 
authors compared systems using two different catalysts bound to the 
same bead and systems with the two catalysts on different beads but 
mixed together. It was found that such systems could be easily re- 
cycled and there was little interference of one catalyst on another. 
These dual catalyst systems can be employed in one "pot" to perform 
multistep organic syntheses with intermediate purification steps 
eliminated. 

Although there are several disadvantages to this heterogenizing 
of homogeneous catalysts, there are many advantages. The major draw- 
backs are: side reactions resulting in the formation of surface 
functional grouos other than those desired; the increased complexity 
of the polymeric ligand; and the difficulty in characterizing the 
active sites. Some of the advantages are: easy separation from the 
reaction mixture so there is virtually no product contamination by 



47 

spent catalyst; reuse without loss of activity; great enhancement 
in product selectivity; less sensitivity to poisoning impurities; 
increased activity in some cases; and a great versatility since 
the polymer support can be easily modified. 

REFERENCES 

1. 0. F. Joklik, Chem. Eng. , 80, ^9 (Oct. 8, 1973). 

2. Z. M. Michalska and D. E. Webster, Chem. Technol . . 5, 117 (1975). 

3. J. C. Bailar, Cat. Rev. Sci. Eng . . 10, 17 (197*0. 
k. R. H. Grubbs, Strem' Chemiker . IV(l), 3 (1976). 

5. C. C. Leznoff, Chem. S oc . Rev . . 2, &5 (197*0. 

6. C. U. Pittman and G. 0. Evans, Chem. Technol . . 3, 5^0 (1973). 

7. L. D. Rollman, Ino rg. Chim. Acta . . 6 137 (1972). 

8. R. J. Card and D. C. Neckers , J. Amer. Chem. Soc > 9Q. 7733 (1977). 

9. J. P. Collman and C. A. Reed, J. Amer. Chem. Soc , £5, 20^8 (1973). 

10. L. R. Melby, J. Amer. Chem. Soc . . 9£, kObk (1975). 

11. J. H. Gaul, Ph.D. Thesis, University of Illinois, 1978. 

12. R. H. Grubbs and L. C. Kroll , J. Amer. Chem. Soc . . £3, 3062 (1971). 

13. J. A. Osborn, F. H. Jardine , J. F. Young, and G. Wilkinson, 
J. Chem. Soc. (A) , 1711 (1966). 

1^. J. M. Moreto, J. Albaiges, and F. Camos , Catalysis (Proceedings 
of the International Symoosium on the Relations between Hetero- 
geneous and Homogeneous Catalytic Phenomena, held in Brussels, 
October 23-2*4-, 197*0 (B. Delmon and G. Jannes, eds.), Amsterdam, 
1975. PP. 339-3^8. 

15. R. H. Grubbs, L. C. Kroll, and E. M, Sweet, J. Molecul. Sci.- - 
Chem . , A7(5), 10^7 (1973)- 

16. R. H. Grubbs and E. M. Sweet, J. Molecul. Catal . . 2, 259 (1978). 

17. J. Reed, P. Eisenberger, B. K. Teo, and B. M. Kincaid, J, Amer , 
Chem. Soc . . p_Q, 5217 (1977). 

18. J. E. Bercaw, R. H. Marvich, L. G. Bell, and H. H. Brintzinger, 
J. Amer. Chem. Soc . . 9j£, 1219 (1972). 

19. G. P. Pez and S. C. Kwan, J. Amer. Chem. Soc . . £8, 8079 (1976). 

20. J. N. Armor, Inorg. Chem .. 12, 203 (1978). 

21. R. H. Grubbs, C. Gibbons, L. C. Kroll, W. 'D. Bonds, and C. H. 
Brubaker, J. Aner. Chem. Soc . . £5, 2373 (1973). 

22. E. Chandrasekaran, Ph.D. Thesis, Michigan State University, 1975. 






48 

23. E. S. Chandrasekaran, R. H. Grubbs, and C. H. Brubaker, 
J. Organomet. Chem . , 120 , 49 (1976). 

24. W. H. Bonds, C. H. Brubaker, E. S. Chandrasekaran, C. Gibbons, 
R. H. Grubbs, and L. C. Kroll, J. Amer. Chem. Soc , 97 , 

2128 (1975). 

25. R. Grubbs, C. P. Lau, R. Cukier, and C. Brubaker, 
J. Amer. Chem. Soc ., 99, 5217 (1977). 

26. C. U. Pittman, L. R. Smith, and S. E. Jacobson, Catalysis 
(Proceedings of the International Symposium on the Relations 
between Heterogeneous and Homogeneous Catalytic Phenomena, 

held in Brussels, October 23-24, 1974) (B. Delmon and G. Jannes, 
eds.), Amsterdam, 1975, pp. 393-405. 

27. C U. Pittman, L. R. Smith, and R. M. Hanes, J. Amer. Chem. Soc , 
97, 1742 (1975). 

28. C. U. Pittman and L. R. Smith, J. Amer. Chem. Soc ., 97, 1749 
(1975). 



49 

THE BINDING OF DI OXYGEN TO METAL CENTERS 

D. Andre d' Avignon April 6, 1978 

The interaction of molecular oxygen with transition metal 
compounds has been of interest for a number of years. Recent re- 
views 1 " 5 sum up a large portion of the work that has been reported. 
Interest in dioxygen compounds has stemmed from their role as oxygen 
carriers in biological systems (e.g., hemoglobins), and their possi- 
ble involvement in homogeneous catalysis oxidation processes. 

This seminar is concerned with dioxygen binding to transition 
metals. Dioxygen complexes are known for most transition elements. 
From a structural viewpoint, these complexes fall into two categories. 
One consists of end-on binding of 2 ; the other group is characterized 
by a sideways-binding of 2 to the metal. Due to the nature of the 
metal dioxygen linkage, the end-on complexes are called superoxo 2 ~ 
compounds, while the sideways-bound species are referred to as peroxo 
2 p - compounds. Both monomeric 1:1 (i.e., metal: 2 stoichiometry) 
and dimeric 2:1 species are known. The connectivity in the 2:1 com- 
plexes is of the form M-O-O-M. 

A wide range of physical techniques have been employed to investi- 
gate structure and bonding in the metal-dioxygen systems. The complexes 
of both types have been structurally characterized by X-ray analysis. 6-7 
I. R. -Raman studies 8 * 9 have also been helpful in characterizing the com- 
plexes, particularly in terms of v and v characteristic absorp- 
tions. The electronic structure 01 the complexes have been studied 
through EPR, 10 * 11 > 12 Vis-UV, 13 X-ray photoelectron spectroscopy, 14 and 
M5ssbauer spectroscopy. 15 Contradictory interpretations concerning 
the nature of M-0 2 bond have developed as a result of the spectral 
studies. 

Recent theoretical work by Goddard employing ab initio GVB 16 
calculations, by Veillard, using ab initio LCAO-MO-SCF 1 ' type calcu- 
lations and also work by Karplus 1 ' B ~using SCF-CI and Xa calculations 
has shed further light on the nature of the dioxygen-metal interaction. 

References : 

1. J. Valentine, Chem. Rev . , 75, 235 (1973). 

2. G. Henrici-Olive and S. Olive, Angew. Chem. Int. Ed. Engl ., 13, 

29 (1974). 

3. F. Basolo, B. M. Hoffman and J. A. Ibers, Accts. Chem. Res ., 8, 

384 (1975). 

4. Lauri Vaska, Accts. Chem. Res ., 9, 175 (1976). 

5. G. McLendon and A. E. Martell, Coord. Chem. Rev ., 19 , 1 (1976). 

6. M. J. Nolte. E. Singleton and M. Laing, J. Am. Chem. Soc . , 97, 

6396 (1975). 

7. L. D. Brown and K. N. Raymond, Inorg. Chem ., 14, 2595 (1975). 

8. T. Shibahara, J. Chem. Soc , Chem. Commun . , 80T (1973). 

9. H. Ruber, W. KlotzbiAcher, G. A. Ozin and A. Vander Voet, 

Can. J. Chem., 51, 2722 (1973). 



50 



10. B. S. Tovrog, D. J. Kitko and R. S. Drago, J. Am. Chem. Soc. , 
98, 51^4 (1976). 

11. B. M. Hoffman, D. L. Diemente and F. Basolo, J. Am. Chem. Soc ., 
92, 61 (1970). 

12. D. Getz, E. Melamud, B. L. Silver and Z. Dori, J. Am. Chem. Soc ., 
97, 3846 (1975). 

13. V. M. M^sdowski, J. L. Robbins, I. M. Treitel and H. B. Gray, 
Inorg. Chem ., 14, 2318 (1975). 

14. J. H. Burness, J. G. Dilla rd and L. T. Taylor, J. Am. Chem. Soc , 

97, 6080 (1975). 

15. K. Snartalian, G. Lang, J. P. Collman, R. R. Gagne and C. A. Reed, 
J. Chem. Phys ., 63, 5375 (1975). 

16. B. D. Olafson and ¥. A. Goddard, III, Proc. Natl. Acad. Scl . 
U.S.A ., 7i, 1315 (1977). 

17. A. Dedien, M. M. Rohmer and A. Veillard, J. Am. Chem. Soc , 

98, 5789 (1976). 

18. B. H. Huynh, D. A. Case and M. Karplus, J. Am. Chem. Soc ., 99 , 
6103 (1977). 



51 



SOME ASPECTS OF THE CHEMISTRY OF HYDRIDO BIS CYC LO PENTAD I ENYL METAL 

COMPOUNDS 

Greg Pearson April 13, 1978 

Transition metal hydrides of the general formula Cp 2 MH undergo 
a variety of reactions with olefins and acetylenes. The nature of 
the reaction is determined by the degree of electronic and coordin- 
ative saturation of the metal. 

Cp 2 ZrH 2 readily reacts with electron donating olefins, however, 
insertion products have never been isolated due to the required re- 
action conditions. 1 Using Cp 2 ZrHCl, a series of stable Zr-alkyl 
compounds have been obtained. 2 Once formed, these alkyl compounds 
will incorporate CO to give the corresponding acyl. 3 Reactions of 
both hydrides with acetylenes have yielded a wide variety of alkenyl 
complexes. 4 

Cp 2 MoH 2 and Cp 2 "WH 2 are both electron rich and display pronounced 
basicity. s > & > 7 A charge transfer interaction is the initial reaction 
between these hydrides and strong Lewis acids. Continued reaction 
leads to insertion into the metal-hydrogen bond. 8 -' 3 > 10 > 12 -> 12 

Cp 2 NbH 3 and Cp 2 TaH 3 are both coordinatively saturated, with all 
available orbitals used in bonding. Therefore, a mechanism is re- 
quired which opens the required coordination site. The mechanism 
demonstrated is the elimination of dihydrogen to produce the reactive 
intermediates [Cp 2 NbH] and [Cp 2 TaH]. 13 These have been found not 
only to coordinate and react with olefins and acetylenes but also to 
catalyze the H/D exchange between dihydrogen and benzene. 13 ' 14 * l5 

References 

1. P. C. Wailes, H. Weigold, A. P. Bell, J. Organomet. Chem . , 45, 
C32 (1972). 

2. D. W. Hart, J. Schwartz, J. Am. Chem . Soc . , 96, 8115 (1974). 

3. C. A. Bertelo, J. Schwartz, J. Am. Chem . Soc . , 97 , 228 (1975). 

4. P. C. Wailes, H. Weigold, A. P. Bell, J. Organomet . Chem . , 27 , 
373 (1971). 

5. J. C Green, S. E. Jackson, B. Higginson, J.C.S . Dalton , 
403 (1975). 

6. A. Nakamura, S. Otsuka, J. Am. Chem . Soc . , 95 , 5091 (1973). 

7. J. C. Green, M. L. H. Green, C. K. Pront, J.C.S . Chem . Comm. , 
421 (1972). 

8. A. Nakamura, S. Otsuka, J. Am. Chem . Soc . , 95, 72o2 (1973). 

9. A. Nakamura, S. Otsaka, J. Am. Chem . Soc . , 9o, 3456 (1974). 

10. A. Nakamura, S. Otsuka, J. Am. Chem. Soc . , gT , 1886 (1972). 

11. A. Nakamura, S. Otsuka, J. Mole . Catal., 1, 285 (1975/76). 

12. A. Nakamura, K. Doi, K. Tatsumi, S. Otsuka, J. Mole . Catal . , 
1, 417 (1975/76). 

13. F. N. Tebbe, G. W. Parshall, J. Am. Chem . Soc , 93, 3793 (1971). 

14. J. A. Labinger, J. Schwartz, J. Am. Chem . Soc . , 97, 1596 (1975). 

15. U. Klabunde, G. W. Parshall, J. Am. Chem. Soc., 9^, 908l (1972). 



52 



APPLICATIONS OF PHASE TRANSFER CATALYSIS TO ORGANOMETALLIC 

SYSTEMS 
Paul C. Adair April 20, 1978 

Phase transfer conditions consist of a two-phase system of 
aqueous base and an immiscible organic solvent . 1 The catalyst, which 
resides primarily in the aqueous phase, commonly is either a crown 
ether binding an alkali metal cation or a tetraalkyl ammonium halide 
salt. Ion pair migration to the organic phase yields very reactive 
species which may react with the target substrate. This technique 
was initially utilized in organic chemistry in 1965.2 Since that 
time, the catalysis of organic reactions by phase transfer has 
exhibited a phenomenal growth. A few examples of reactions facilitated 
by such means are carbene reactions, nucleophilic substitutions, 
alkylations of ketones and nitriles , and formation of ethers and 
esters . 3 

The first application of phase transfer catalysis to a system 
containing a transition metal was published in 1973. 4 Five 1, 1-di- 
chloro-2-f errocenylcyclopropanes were obtained in high yield by the 
generation of dichlorocarbene in the presence of the corresponding 
vinylf errocenes . More recently, several papers have reported 
organic reactions which are catalyzed or facilitated by metal carbonyls 
under phase transfer conditions. Formation of ketones or diketones 
from bromoketones 5 and synthesis of carboxylic acid derivatives 
from aromatic halides^?? have occurred in the presence of Co2(CO)8- 
Aromatic nitro compounds may be reduced to the corresponding amines 
by Fe3(CO)]_2- ' In each of these cases, it is believed that 
metal carbonyl anions are produced under phase transfer conditions. 
These anions react with the organic species in the nonaqueous phase 
to yield the observed products. 

Phase transfer catalysis has also been useful in the synthesis 
of metal complexes. High yield syntheses of cobalt carbonyl compounds 
derived from Co2(CO)3 have been reported. ^ Use of allyl bromide 
led to ir-allylcobalt carbonyls and use of trihalomethyl compounds led 
to the alkylidynetricobalt nonacarbonyls . By the reaction of 
thiobenzophenones with Feq(C0)i2 an interesting ortho-metallated 
complex may be produced. ■*- The relatively inert group VI metal 
carbonyls (M(C0)6; M=Cr, Mo, W) have been shown to undergo accelerated 
substitution by tertiary phosphines , triphenylarsine , or dipyridyl 
in the presence of an aqueous sodium hydroxide/benzene system with 
tetra-n-butylammonium iodide as the phase tranfer catalyst. ^ 
Neutral group VI and iron complexes have been shown to undergo -*-°0 
labeling of the carbonyl ligands under similar conditions when the 
aqueous layer is H2 0.^-3 Again, the role of the phase transfer 
catalyst in each of these reactions apparently is to form a metal 
carbonyl anion, which then undergoes further reaction. 



53 



Phase transfer catalysis has been shown to be applicable for 
organometallic reaction systems that involve anionic intermediates. 
In many uses it is preferred to other routes, offering the 
advantages of optimized yield, reaction time, and other reaction 
parameters. As this technique is only in its infancy, it will 
undoubtedly become increasingly important in the future. 

References 

1. E. Dehmlow*, Angew. Chem. Int. Ed., 13, 170 (1970). 

2. M. Makoza and B. Seraf inowa , Rocz. Chem., 3_9, 1223, 1401, 1595, 
1799, 1805 (1965). 

3. E. Dehmlow, Angew. Chem. Int. Ed., 16, 493 (1977). 

4. W. Weber, J. Shepard and G. Gokel , J. Org. Chem., 3j3_, 1913 (1973). 

5. H. Alper and K. Logbo , Tet. Lett., 3J_, 2861 (1977). 

6. H. Alper and H. des Abbayes, J. Organomet . Chem., 134 , Cll (1977). 

7. L. Cassar and M. Foa , J. Organomet. Chem., 134 , C15 (1977). 

8. H. Alper, D. DesRoches and H. des Abbayes, Angew. Chem. Int. Ed., 
16, 41 (1977). 

9. H. Alper and H. des Abbayes, J. Am. Chem. Soc, 9_9, 98 (1977). 

.0. H. Alper, H. des Abbayes and D. DesRoches, J. Organomet. Chem., 121 
C31 (1976). 

H. Alper and D. DesRoches, J. Organomet. Chem., 117 , C44 (1976). 
B. Shaw and K. Hui , J. Organomet. Chem., 124 , 2 62 (1976). 
D. Darensbourg and J. Froelich, J. Am. Chem. Soc, 100 , 338 (1978). 



ELECTRON DYNAMICS IN BINUCLEAR COPPER(ll) COMPLEXES 
U1ITH EXTENDED BRIDGING GROUPS 



54 



Timothy R. Felthouse 
Introduction 



(Final Seminar) 



April 27, 1978 



The 
current 
a weak 
centers 
this in 
Hamilto 
which d 
materia 
lattice 
organic 
complex 
meter a 
work ha 
of a po 
action 
given t 
paramet 
bridge 



study 
chemic 
couplin" 
, is kn 
teracti 
nian H 
evelop 
Is exhi 
polyme 
biradi 
es has 
nd the 
s been 
lyatomi 
between 
o compl 
er on t 
orbital 



of elect 

al resea 
g betwee 
own as a 
on is ga 

= ~ 2J §1 

between 

bit exch 
rs, tran 
cals . 
shown a 
Cu-O-Cu 
to estab 
c bridgi 
two cop 
exes whi 
he coppe 
s capabl 



ron- 

rch . 

n th 

mag 

uged 

•S 
-2, 

van 

ange 

siti 

ecen 

line 

brid 

lish 

ng 

pe 

ch w 

r(ll 

e of 



' ? 



dynamica 

One of 

e electr 

netic ex 

by the 
and this 
ous coup 

interac 
on metal 
t work o 
ar relat 
ging ang 

various 
roup to 
II) ions 
ill eluc 
) single 

propaga 



1 ph 

the 
ons 
chan 
exch 

par 
led 
tion 

clu 
n di 
ions 
le. 1 

cri 
supp 
. P 
idat 
-ion 
ting 



enomena permeates many areas of 
se phenomena, characterized by 
from two or more paramagnetic 
ge interaction. The strength of 
ange parameter J in the spin 
ameter measures the splittings 
electronic states, A variety of 
s, some of which include extended 
sters in metalloproteins , and 
-(a-hydroxo-bridged copper(ll) 
hip between the exchange para- 
One objective of the present 
teria for judging the viability 
ort a magnetic exchange inter- 
articular attention has been 
e the dependence of the exchange 
ground state and the relative 
the exchange interaction. 



Oxalate-Bridoed Comolexes 



A serie 
having the 
dien [hn(CH 

N(Ch0 2 ) 2 ]. 
or PFg- I 

inflexible 
are pyramid 
idal (TBP) 
complexes w 
reveals tha 
copper( II ) 
per( II ) gro 

a magnetica 
such, coppe 
system are 
mediated by 
orbital (MO 
interaction 



s of c 

genera 

2 CH 2 NH 

or Et 

n thes 

bridoi 
al (SP 
in [Cu 
ith va 
t the 
ground 
und st 
lly di 
r hype 
observ 

the 
) anal 

as th 



ompl 
1 fo 

2 ) 2 ] 

t-die 
e co 
ng g 
) in 
2 (Et 
r iab 
exch 
sta 
ate 
lute 
rf in 
ed i 
-C-0 
ysis 
e pr 



exes 
rmul 
i dp 
n [N 
mple 
roup 

[Cu 
cdie 
le-t 
ange 
te ( 
( Ul 

env 

e sp 

n th 

por 
3,4 

inci 



bri 
a [C 

t [h 

C g H 5 
xes 
as 
2 (di 

n) 2 ( 

empe 
int 
J = 
< 
iron 
litt 
e pu 
tion 
indi 
pal 



dged 

u ? (" 

!\ltCH 

(CH ? 

the z 

the 

en) ? 

c 2 o 4 

ratu 
erac 
-37. 
.5 c 
ment 
ings 
re s 

of 
cate 
supe 



by t 

dien" 
oCHoC 
CH ? M( 
oxala 
coppe 

( )» 

re ma 
tion 
4 cm" 
m" T ). 
for 
appr 
olid, 
the o 
sap 
rexch 



he ox 

) 2 (c 2 

H 2 ^H 2 
CoH.) 
t§ 8i 

r(H) 
)J(C1 
h 4 ) 2 . 
gneti 
is mu 
1) th 

The 
the b 
opria 

The 
xalat 
rimar 
ange 



alate 

n 

anion 
geom 

^kl 

c sus 
ch gr 
an f o 

BPh 
inucl 
te f o 

exch 
e bri 
ily 
pathw 



dianio 
Y) 2 , wh 
Me^dien 
and Y"i 

serves 
etry ch 

to tri 
aracter 
ceptibi 
eater f 
r the S 
count 
ear cat 
r an ex 
ange in 
dge and 
-0 thro 
ay. 



n were 
ere "d 

[nch 3 

s BPh, 
4 
as a 

anges 

gonal 

izatio 

lity a 

or the 

P d x2 _ 

erion 

ions a 

change 

teract 

a mol 

ugh-sp 



pre 
ien" 
(CH 2 

, C 
rela 
from 
bipy 
n of 
nd E 
TBP 
2 c 
rov 
nd , 
-cou 
ion 
ecul 
ace 



I 



pared 

is 
CH 2 - 

io -, 

t i V e 1 y 

squ- 
ram- 

the 
PR 

d z 2 
op- 
ides 
as 

pled 
is 
ar 



Azide-Bridqed Complexes 



Replacement of the oxalate dianion with two azide ions produced 
complexes of the form [Cu 2 ( "dien" ) 2 (N 3 ) 2 ]( Y ) 2 . The magnetic suscept- 
ibility and EPR data substantiate the dimeric formulation and again 
the complexes having the Et5dien ligand which enforces a TBP geometry 



55 



showed the largest antif erromagnet ic exchange interaction. The brid- 
ging mode of the azide ion could not be inferred from various spectro- 
scopic techniques, and the X-band EPR spectra revealed unusually large 
zero-field splittings. X-ray structural analysis of [Cuolf'ierdienjn- 
^3^2^^ Pn 4^2 revealed that the azide ions were bridging in a |JL ( 1 , 3 ) 
fashion. The zero-field splittings greatly exceed the values calcu- 
lated from purely dipolar contributions and arise from pseudodipolar 
interactions which have their origin in exchange interactions in ex- 
cited states. 



Aromatic Diamine-Bridqed Complexes 



Th 

d 2 gr 
plexes 
( tren) 
sents 
dine ( 
the ma 
intera 
and -3 
comple 
In ord 
x-ray 
discre 
of mor 
dihedr 
lation 
work o 
group 



e ef f e 
ound s 

with 
2 (DA)] 
an aro 
BZD), 
gnet ic 
ct ions 
5.1 cm 
xes ex 
er to 
struct 
te TBP 
e than 
al ang 
s the 
f the 
cannot 



-f 



ctiv 
tate 
more 
(Y) 4 
mati 
and 

sus 
or 
d 

hibi 
conf 
ure 

bin 

12 
les 
prin 
BZD 

dis 



enes 

in 

ext 

wer 
c di 
Y~ i 
cept 

the 
epen 
ted 
irm 
of [ 
ucle 

%. 

and 
cipa 
mole 
coun 



s in p 

a TBP 
ended 
e prep 
amine 
s vari 
ibilit 
PPD-b 
ding u 
exchan 
the in 

Cuo( tr 
ar cat 
The ph 
are co 
1 supe 
cule a 
t a it 



ropa 

geom 

brid 

ared 

such 

ousl 

y da 

ridg 

pon 

ge c 

tram 

en) 2 

ions 

eny 1 

nnec 

rexc 

ltho 

spin 



gati 
etry 
ging 
, wh 

as 
y NO 
ta r 
ed c 
the 
oupl 
olec 
(BZD 

and 

rin 
ted 
hang 
ugh 

pol 



ng a 
led 
gro 

ere 

p-ph 

3~» 
evea 

ompl 

coun 

ing 

ular 

)](N 

nit 
gs i 
by a 
e pa 
the 
ariz 



n exch 
to at 
ups . 
tren i 
eny ien 
ClO^-, 
led re 
exes ( 
terion 
( J var 
natur 

rate a 
n the 

C-C s 
thway 
low sy 
ation 



ange interaction o 
tempts to prepare 
Complexes of the f 
s N(CH 2 CH2MH2) 3 , D 
ediamine (PPD) or 

or Pfg""» Remarka 
latively large exc 
J varies between - 
) and even the BZD 
ies from -3.3 to - 
e of the interacti 
as solved^ and it 
nions with a Cu-Cu 
BZD bridge have no 
ingle bond. From 
is by means of the 
mmetry of the brid 
contribution • 



f the 

com- 

orm £Cu2- 

A repre- 

benzi- 

bly, 

hange 

19.8 cm" 1 

-bridged 

4.5 cm"" 1 ) . 

ons , the 

reveals 

distance 
nzero 
WO calcu- 

0* frame- 
ging 



Conclus ions 



T 
magne 
one m 
molec 
with 
to pr 
ilar 
act iv 
elect 
relat 
rate 
coura 
dynam 



he pr 
t ic m 
eans 
ular 
those 
oduce 
s i t u a 
e. T 
ron t 
ively 
of el 
ge f u 
ical 



esen 
etal 
of d 
elec 

of 

a d 
tion 
his 
rans 

Ion 
ectr 
rthe 
phen 



ce o 
ion 
eter 
tron 
the 
eloc 
for 
is e 
f er 
g li 
on t 
r ex 
omen 



f a ma 
s sepa 
mining 

trans 
bridge 
alized 

two m 
specia 
reacti 
f et ime 
ransf e 
perime 
a and 



gnet 
rate 

if 
f er. 

whi 

ele 
etal 
lly 
on i 

of 
r . 

ntal 
thei 



ic e 
d by 
that 
Th 
ch h 
ctro 

ion 
true 
s in 
the 
It i 

and 
r in 



xchan 

an e 

brid 

e int 

ave t 

nic s 

s wh i 

when 

the 

precu 

s hop 

theo 

terre 



ge in 
xtend 
ging 
erf ac 
he pr 
truct 
ch ar 

the 
nonad 
rsor 
ed th 
retic 
latio 



teract 
ed mol 
group 
e of t 
oper s 
ure is 
e exch 
precur 
iabat i 
is f ou 
at the 
al wor 
nships 



ion betw 
ecular u 
could su 
he metal 
ymmetry 

a quali 
ange c u 
sor comp 
c regime 
nd relat 
se studi 
k involv 



een two para- 
nit provides 
pport intra- 

ion orbitals 
and energy 
tatively sim- 
pled or redox 
lex in an 
» i.e., a 
ive to the 
es will en- 
ing electron- 



References 



1. V. H. Crawford, H. Ul . Richardson, J. R. Was son, D. J. Hodgson, and 
W. E. Hatfield, Inorq . Chem ., 15, 2107(1976). 



56 



2. N . F. Curtis, I. R. N. WcCormack, and T. N. Waters, ^J. Chem . Soc . , 
Dalton Trans ,, 1537(1973). 

3. T. R. Felthouse, E. J. Laskowski, and D. l\l . Hendrickson, Inorq . 
Chem ., 16, 1077(1977). 

4. P. J. Hay, J. C. Thibeault, and R. Hoffmann, _J. Ajm. Chem . Soc . , 97 , 
4884(1975). , 

5. T. R. Felthouse and D. N. Hendrickson, Inorq . Chem . , 17 , 444(1978). 

6. T. R. Felthouse, E. N. Duesler, and D. N. Hendrickson, J_. Aim. Chem . 
Soc, 100 , 618(1978) . 

7. H. Taube in "Bioinorganic Chemistry II", K. N. Raymond, Ed., Ad- 
vances in Chemistry Series 162, American Chemical Society, 1977, pp 
127-144. 



- 59 - 

ELECTRON EXCHANGE ACROSS CYANIDE AND CYANID3-LIKS 

BRIDGES IN COPPER (II) SYSTEMS 

David S. Bieksza Final Seminar 18 May 1973 

Paramagnetic centers in transition natal cluster compounds nay interact 
with each other by means of derealization of unpaired ©lectron density from one 
r.etal site to another — electron exchange. The interaction arises either from 
direct overlap of th«> metal orbitals, forming metal-metal bonds, or from super- 
•::::change through diamagnetic bridge orbitals. This derealization causes ferro- 
magnetic or antiferromagnetic bulk magnetic behavior. Ferromagnetism arises 
from the spin alignment of unpaired electrons which occupy orthogonal orbitals, 
Antif erromagnetism arises from spin pairing when unpaired electrons occupy the 
same orbital. In either case the coupled spin states are split; for binuclear 
copper (II) systems a singlet state and a triplet state result. Describing the 
exchange interaction with a spin Hamiltonian of the form H = -2J SjjSg, the 
states are split by an amount equal to 2J. Variable temperature magnetic sus- 
ceptibility experiments can determine both the sign and magnitude of J, the 
exchange parameter. 

Cyanide-bridged dircers were prepared of the form (Cu2Lj^CN)(FF^)o, where 
L is either 2,2" -bi pyridine or 1,10-phenanthroline, and (Cu2tren2CN)(PF£)'3 t where 
tren is 2,2* ,2* •-triarainotriethylamine. These ligands were chosen on the basis 
that all three tend to enforce trigonal bipyramidal geometry about copper (II) » 
as well as their ability to stabilize the metal from reduction to cuprous cyanide. 
Infrared spectra indicated end-to-end bridging by the cyanide in each case. In 
the SPR spectra only isotropic signals around g=2»i were obtained, the result of 
exchange averaging between copper centers with misaligned g- tensor axes. Magnetic 
susceptibility found the exchange interaction to be antif erromagnetic for each 
dimer. The exchange parameters were measured to be -9»^ cm"" 1 for the bipyridine 
compound, -29 cm"" 1 for the phenanthroline compound, and -88 cm"" 1 for the tren 
compound. The variation in J values arose from the varying amounts of overlap 
of the cyanide sigxaa orbitals with the copper do orbitals. 

To determine how extended the bridge could become and still support an 
exchange interaction via a sigma-only pathway, dicyananiide-bridged dimer3 were 
prepared of the form (Cu2tren2N(CN)2) Y3, where Y is B?V", C101*", PF6~, or 
B(C6H5)"~. Magnetic susceptibility found no interaction (1JI less than 0.5 cm"" 1 ) 
for all compounds except the tetrapheny lb orate dimer, where J =2, 5 cm" 1 . The 
pattern was repeated in the EPR spectra: all the compounds had an isotropic 
derivative around g=2.2 except for the tetraphenylborate dimer, which had an 
-•ixial spectrum. Likewise the IR and Raman spectra of the three non-interacting 
compounds matched each other in the cyanide region, but the spectrum of the fourth 
compound had different band positions and intensities. The explanation for these 
observations centered on the angle at th_s middle nitrogen of the dicyanamide 
bridge, ordinarily 120-130°. A larger angle, about 160°, was consistent with the 
EPR, IR and Raman data. Molecular orbital calculations indicated that as the 
angle increased overlaps both within the bridge and botween bridge and metal 
orbitals increase, and so a more viable pathway for electron exchange was 
created. 



- 60 - 

PART ONE: BINUCLEAR PYRAZ INE-BRIDGED COPPER(II) AND VANADYL COMPLEXES 

PART TWO: CONSEQUENCES OF HYDROGEN BONDING INTERACTIONS 

FOR COBALT-DIOXYGEN COMPLEXES 

J. Patrick Cannady Final Seminar July 11, 1978 

Binuclear tans it ion metal complexes in which pyrazine serves as a 
coordination bridge between the two metal centers have been the subjects 
of much interest in recent years. Determinations of the rate of intra- 
molecular thermal electron transfer in the so-called Creutz and Taube 
mixed-valence ion, L(NH3)cRu-pyz-Ru(NH2)c]], first reported* in 1969, have 
been the focus of much discussion in the literature. In a recent report, 
it was found that the unpaired electron, which is localized in a (d xz »d ) 
orbital on the Ru(III) center, is conveyed by the pyrazine bridge fast on 
the H NMR time scale, but slowly on the EPR timescale. However, the re- 
lated bis-trivalent ion, £(NH 3 ) Ru-pyz-Ru(NH-) ") 6 t gave no evidence of a 
magnetic exchange interaction down to 4.2°K. In several polymeric copper (I I) 
species, in which infinite (Cu-pyz) chains exist, exchange interactions 
have been detected, e.g., for [Cu(N0.02-pyz3 x , J=-7.4 cm" . Other poly- 
meric copper(II)-pyz species, such as [Cu(hfac) 2 pyz3 x and tCuChfac^DABCOj^,, 
where hfac= hexaf luoroacetylacetone and DABC0= 1,4-diazabicyclo 2.2.2 octane, 
have been investigated but show no evidence of a superexchange interaction 
down to 1.8°K. Extended Huckel molecular orbital considerations of Hoffmann 
and coworkers led them to conclude that both pyrazine and DABCO should be 
very capable of serving as QT-type pathways for antif erromagnetic superexchange. 

In an effort to determine the extent to which 0"- and IT- type interac- 
tions could be propagated by these bridging units under the most favorable 
electronic conditions, a series of copper(Il) and vanadyl complexes was 
prepared. A ranee of electronic ground states was obtained' in this series 
by using Cu(tren) (ground state = d_2), Cu(hfac) 7 (ground state = d v 2_ v 2)» 
and VOChfacK (ground state = d x ) as the one-unpaired-electron sources. 
EPR and magnetic studies support a binuclear formalism for these compounds. 
Values of the superexchange parameter, J, were most negative for the com- 
plexes [V0(hfac)2*j!2^» where B = pyrazine (J ■ -19.0 cm ), 2-methylpyrazine 
(J «* -6.7 cm"*), and 2,5-dimethylpyrazine (J » -4.0 cm"*). fc^tre^pyzJCClO^)^ 
and [Cudifac^lo^ABCO gave values for J of -3.2 cm and<0.5 cm"*, respectively, 
Thus, the most favorable situation for superexchange via pyrazine appears 
to arise from the interaction between thefTorbitals of pyrazine and the d xy 
orbitals on the V centers. This is made geometrically plausible by assuming 
an equatorial coordination of the pyrazine in the vanadyl complexes. (T - type 
interactions in these systems are seen to be very weak or not existent. 
Nevertheless, solution studies on V0(hfac)2 2P^ Z * n methylene chloride 
indicate a decrease of almost three orders of magnitude for the equilibrium 
constant for binding the second V0(hfac) ? unit over that for binding the 
first to pyrazine. This marked decrease in the basicity of the pyrazine 
nitrogen provides an example in which it is apparent that the consequnces 
of a weak coupling of the electronic manifolds of the two metal centers 
does not entirely account for the total effect that the two metal cneters 
can have on each other. 



- 61 - 



PART TWO 

Hydrogen bonding has been invoked to explain the enhanced rate of 
formation at ambient temperatures of the binuclear peroxo bridged species, 
LPCoO CoPL, where P= protoporphyrin^ dimethyl ester and L= imidazole, 
in toluene over solutions in which L= N-methylimidazole, which is incapable 
of serving as a hydrogen bonding acid. In order to investigate the gener- 
ality of this effect for other cobalt-dioxygen systems, experiments were 
undertaken on a system which should better lend itself to investigation 
of this type'. CoSMDPT, where SMDPT = N-methyl-S^-bisCsalicylideneamino)- 
bispropylamine, has been shown by EPR to bind dioxygen reversibly at low 
temperatures as well as at ambient temperatures. ' HI and *-"p NMR studies 
on benzene, toluene and methylene chloride solutions of CoSMDPT to which 
TFE, 2,2,2-trif luoroethanol, has been added indicate that TFE interacts 
with CoSMDPT, either by hydrogen bonding to the phenolic oxygens or the 
imine linkage of the ligand. Exposure to dioxygen results in the formation 
of a diamagnetic species. EPR studies show that the room temperature sig- 
nal of the 1:1 Co-O^ adduct is decreased upon addition of TFE to a toluene 
solution of CoSMDPT under oxygen. Low temperature and frozen glass EPR 
parameters do not show much dependence on the presence of TFE. However, 
spectrophotometric measurements of the 1:1 equilibrium constant for the 
binding of dioxygen to CoSMDPT in toluene and methylene chloride with 
and without TFE do show a marked dependence on the presence of TFE. In 
general, K Q increases on going from toluene to methylene chloride. However, 
the presence of TFE in toluene solutions results in a decrease in Kq , while 
in methylene chloride, a marked increase is observed. The interplay between 
solvation and hydrogenbonding effects in these systems is seen to account 
for these behaviors. 



References 

1. C. Creutz and H. Taube, J. Am. Chem . Soc, 91, 3988 (1969). 

2. B. C. Bunker, R. S. Drago, D. N. Hendrickson, R. Richman, and S. L. 
Kessel, J. Am. Chem . Soc , 100 , 3805 (1978). 

3. H. W. Richardson and W. E. Hatfield, J. Am. Chem . Soc , 98, 835 (1976). 

4. H. W. Richardson, J. R. Wasson, and W. E. Hatfield, Inorg . Chem . , 16 , 
484 (1077). 

5. P. J. Hay, J. C. Thibeault, and R. Hoffmann, J. Am. Chem . Soc , 97 , 
4884, (1975). 

6. M. S. Haddad, D. N. Hendrickson, J. P. Cannady, R. S. Drago, and 
D. S. Bieksza, J_. Am. Chem. Soc . , submitted. 

7. D. V. Stynes, H. C. Stynes, J. A. Ibers, and B. R. James, J. Am. Chem . 
Soc , 95, 1142 (1973). 

8. B. S. Tovrog, D. J. Kitko, and R. S. Drago, J. Am. Chem . Soc ,98, 5144 
(1976). 



9. R. H. Niswander and L. T. Taylor, J. Magn . Resonance , 26 , 491 (1977). 



r> 



• 



UNIVERSITY OF ILLINOIS 

TABLE OF CONTENTS 

INORGANIC SEMINAR ABSTRACTS 

1978-1979 



Page 

REACTIVITY OF POLYSTYRENE-BOUND BIPYRIDINE AND THE 

CATALYTIC POTENTIAL OF ITS TRANSITION METAL COMPLEXES - 

Anton El A'mma 1 

INCLUSION BY MOLECULAR CAVITIES: APPLICATIONS - Les Butler 3 

INTERVALENCE ELECTRON TRANSFER IN MIXED VALENCE COMPOUNDS - 

Bruce C. Bunker 6 

PHOTO-OXIDATION OF INORGANIC COMPOUNDS WITH APPLICATION 

TO SOLAR ENERGY - Richard W. Wegman 9 

DYNAMICS OF ELECTRONIC SPIN TRANSITIONS IN IRON "SPIN- 
EQUILIBRIUM" SYSTEMS - Wayne D. Federer 12 

AQUEOUS MANGANESE (III) COMPLEXES WITH OXYGEN DONOR LIGANDS - 
Brenda R. Shaw Suib 15 

THERMOCHROMIC BEHAVIOR OF SELECTED INORGANIC SYSTEMS - 19 

David M. Hamilton, Jr. 

PHYSICAL PROPERTIES OF THE ONE-DIMENSIONAL CONDUCTING 

MATERIAL BaVS 3 : ITINERANCY AND STOICHIOMETRY - Jeff Kelber 22 

SOME RECENT DEVELOPMENTS IN THE INTERCALATION CHEMISTRY OF 

THE LAYERED TRANSITION METAL DICHALCOGENIDES - Deborah E. 

Samkoff 25 

LIGAND EXCHANGE STUDIES OF METHYLCOBALAMIN AND MODEL 
METHYLATOCOBALT (III) CHELATE SYSTEMS - Pamela Milton 28 

NITROGEN-14 NUCLEAR OUADRUPOLE RESONANCE SPECTRA OF 

COORDINATED PYRIDINE - Gerald V. Rubenacker 30 

KINETIC ISOTOPE EFFECTS OF OXIDATIVE ADDITION OF HYDROGEN 

TO METAL COMPLEXES - David J. Blumer 31 

SUPER-IONIC CONDUCTIVITY IN NON-STOICHIOMETRIC OXIDES - 

Stanley A. Roth 33 

POLYOXOANION CHEMISTRY - Peter Dimas 36 



Table of Contents 
Inorganic Seminar Abstracts 
University of Illinois 
Page 2 



ASYMMETRIC HYDROGENATION USING WILKINSON-LIKE CATALYSTS - 

Sharon A. Brawner 39 

FERROFLUIDS - Arrietta Walker 42 

METAL VAPORIZATION: A ROUTE TO SOME NOVEL ORGANO-TRANSITION 

METAL CHEMISTRY - Debra S. Strickland 44 

HOMOGENEOUS CATALYSIS OF THE WATER GAS SHIFT REACTION - 

Robert Olsen 47 

STRUCTURAL STUDIES OF PROTEINS IN SOLUTION USING NMR AND 
PARAMAGNETIC PROBES - James R. Stahlbush 50 

SEMICONDUCTING RARE EARTH METALS - Michael K. Kroeger 5 3 

SUPPORTED METAL CARBONYL CLUSTERS - Alan Zombeck 55 

ELECTRON DYNAMICS OF METAL-QUINONE COMPLEXES - Stephen L. 

Kessel 58 

A MAGNETIC AND MOSSBAUER STUDY OF SOME MIXED VALENCE COMPOUNDS - 
Karen Hassett 61 

THE NATURE OF ACTIVE SITES IN CATALYTIC ZEOLITES - Steven L. 

Suib 6 2 

NUCLEAR QUADRUPOLE RESONANCE STUDIES OF METAL COMPLEXES OF 
BIOLOGICAL INTEREST - Carol I. H. Ashby 6 5 

THE METAL-CARBON TRIPLE BOND - Allen D. Clauss 6 8 

MAGNETIC EXCHANGE INVESTIGATIONS OF TRANSITION METAL COMPLEXES - 
Lynn C. Francesconi 70 

BASE CATALYSIS OF LIGAND SUBSTITUTION IN METAL CARBONYLS - 

Peter A. Bellus 72 

ELECTRON TRANSFER IN MIXED VALENCE METALLOCENES - Jack A. Kramer 74 



REACTIVITY OF POLYSTYRENE- BOUND BIFYRIDINE AND THE CATALYTIC 
POTENTIAL OF ITS TRANSITION METAL COMPLEXES 

Anton El A'mma September 21, 1978 

Since 1971, there has been a growing interest in catalysis 
by solid supported transition metal complexes. 1 The metal complex 
which is grafted onto a support can either be an established 
homogeneous catalyst or its metal center could acquire catalytic 
activity after the process of heterogenization. 

Crosslinked polystyrene is a commonly used solid support and 
numerous ligands have been anchored to this resin through covalent 
bonds. The resin which is functionalized with the analog of tri- 
phenylphosphine 2-4 is known to heterogenize potent homogeneous 
catalysts. However it has the following drawbacks: 1, air 
sensitivity of the phosphorus center; 2, lability of phosphine- 
metal linkage in some systems; 3, polymer crosslinking induced 
by the complex which modifies the primary structure of the support. 

In order to bypass the above disadvantages, we have indepen- 
dently undertaken a study of the reactivity of a polymer bound 
bipyridine [(P)-Bipy] reported by Neckers et al. 5 and investigated 
the catalytic activity of some of its transition metal complexes 
towards hydrogenation. The choice of immobilized bipyridine steins 
from the fact that this ligand stabilizes a wide range of metal 
oxidation states and functions as a chelate. 

.The bonding properties of (p)-Bipy were studied by reacting 
it with a series of carbonyl containing metal complexes and then 
comparing the IR spectra (C-0 stretching region) of the various 
polymer attached products with those of compounds derived when 
bipyridine undergoes the corresponding reactions. Polymer bound 
pyridine was used in similar reactions to ascertain the bidentate 
behavior of (p)-Bipy. • Thus, (p)-Biny was reacted with Mo(C0) c * 
(CO)pRhCl/?, and (CO) (PPh 3 )RhCl/2 to form (p) -BinyMo(co) 4 , 
[(P)-BipyRh(C0) 2 Cl, (P)-BipyRhCOCl] and (p)-BipyRh(PPh 3 )C1 res- 
pectively. The further reaction of (p)»BipyMo(C0) 4 with PPh 3 
generated (p)-BipyMo(co) 3 (PPh 3 ) . These investigations indicate that 
(p)-Bipy functions as a bidentate ligand. 

Catalytic hydrogenations were observed with the following 
catalyst precursors: (P)-BipyPtCl 2 , (P)-BipyPdCl 2 , polystyrene 
dispersed palladium metal and (P)-BipyRh(PPh 3 )C1. The system 
(P)-BipyPtCl 2 shows a high degree of selectivity towards the 
hydrogenation of terminal olefinic bonds. The material 
(P)-BipyPdCl 2 reduces a wide variety of coordinating and non- 
coordinating substrates (with isomerizaticn where applicable) 
and polymerizes methyl acrylate. The activity of this catalyst 
can be attributed in part to the presence of metallic palladium. 
Polystyrene dispersed palladium was found to be a factor of 
2-3 times slower (rate of H 2 uptake) than r j% Pd/C in hydrcgenating 
p-benzoquinone and nitrobenzene. This system (P )-BipyPh(P?h 3 ) CI 



is even more ef . ,ive in hydrogenating internal and terminal 
double bonds th; (P)-BipyPdCl 2 . Tentative mechanisms are dis- 
cussed for the activity of both (P)-BipyPtCl 2 and (p)-BipyRh(PPh 3 )ci. 
All of the abo\ e catalyst precursors have been reused without any 
loss of activity. 

In summary, the use of immobilized bipyridine has facilitated 
the understanding of the coordination sphere of the grafted metal 
complex and generated some novel hydroge nation catalysts. 



References : 

1. J. C. Bailar, Cat. Rev. Sci. Eng. , 10, 17 (197*0. 

2. R. H. Grubbs and L. C. Kroll, J. Am. Chern. Soc., 93, 3062 (l97l). 

3. R. H. Grubbs, L. C. Kroll and E. M. Sweet, J. Mol. Sci. Chem. , 
A7(5) , 1047 (1973). 

4. J. P. Collman, L. S. Hegedus, M. P. Cooke, J. R. Norton, 
G. Dolcetti, and D. N. Marquardt, J. Am. Chem. Soc., 94, 
1789 (1972), 

5. R. J. Card and D. C. Neckers, J. Am. Chem. Soc., 99, 7733 (1977). 






INCLUSION BY MOLECULAR CAVITIES: APPLICATIONS 
Les Butler October 12, 1978 

The flexibility offered by the use of molecular cavities has 
excited much interest in spite of the added complications involved 
with their application. Zeolites, 1 silicates, 2 and graphite 3 
represent long-standing examples of exploitation of cavities for 
various applications. The quest for solution phase cavities is 
• satisfied, at least in part, with cyclodextrins , H ' 5 sugar deriva- 
tives, and cryptands, 6 synthetic multicyclic heteroatom compounds. 
The purpose of this seminar is to review some properties and sel- 
ected examples of the use of molecular cavities, some of which are 
large enough to contain substituted aromatics and others which can 
only include small molecules and metal ions. 

SMALL CAVITIES 

Most of the research with cryptands has dealt with the bi- 
cyclic compounds such as N (CH2CH2OCH2CH2OCH2CH 2 ) 3N denoted as 
[2.2.2], although other topologically different derivatives are 
known. Early research showed that [2.2.2] prefers to include 
the potassium cation while [3.3.2] exhibited nearly equal for- 
mation constants for potassium and rubidium cations. The observed 
stability was described in terms of the number and type of binding 
sites leading to the "cryptate effect" while the selectivity was 
seen to be a function of the fit of cavity size with the metal 
cation size. 8 The temperature dependence of the complex formation 
showed that the change in enthalpy reflected the selectivity of 
a particular cryptand. 9 In methanol, both selectivity and stability 
of the included complexes were enhanced. 8 ' 9 It was observed that, 
in methanol, the alkaline-earth complex with the greatest stability 
also possessed the greatest activation enthalpy for dissociation. l 
Similar characteristics were also observed with alkali cations 11 
and it was suggested that the transition state for formation of the 
complex closely resembles the state of the reactants. 10 ' 1 1 The 
energy of desolvation of the metal cation would be compensated by 
interaction with the complex. 

Recent applications of small cavities have included the 
preparation of inert lanthanide cryptates , l 2 stabilization of 
the sodium anion, 13 ' 14 and use in forming isolated counter-ions. 15 

LARGE CAVITIES 

A convenient source of large cavities is provided by the 
cyclodextrins which are a-l,4-linked cyclic oligomers of D- 
glucopyranose (a, £, and y refer to 6 , 7 and 8 glucose units, 
respectively) which form a hollow, hydrophobic, truncated cone 
ringed by twelve secondary hydroxy 1 functions at the large end, 
and six primary hydroxyl functions at the other end. 16 A wide 
variety of molecules can be included into cyclodextrins resulting 
in many suggestions for the driving force for inclusion. Nuclear 
magnetic resonance 17 and crystal structure determinations 18 have 
been employed to examine the nature of the binding and orientation 
of included benzoic acid and sodium benzoate. These studies have 



shown that in both cases the acid function is located inside the 
cavity. A comprehensive model of the thermodynamics of inclusion 
of aquated apolar substrates was recently reported. 19 

The applications of cyclodextrins have been numerous: the 
resolution of chiral sulfinyl compounds, 20 the preparation of 
Vitamins Ki and K 2 , 21 and catalytic hydrolysis of amides. 22 
Furthermore, the substrate binding by enzymes can be modeled 
with cyclodextrins or derivatives. 23 ' 21 * 

References 

1. P. A. Risbo d and D. M. Ruthven, J. Am. Chem. Soc, 100 , 
4919 (1978). 

2. J. M. Adams, J. A. Ballantine, S. H. Graham, R. A. Laub, J. H. 
Purnell, P. I. Reid, W. Y. M. Shaman and J. M. Thomas, Angew. 
Chem. Int. Ed. Engl., 17, 282 (1978). 

3. J. E. Fischer and T. E. Thompson, Physics Today, July 36, 1978. 

4. M. L. Bender and P. W. Griffiths, Adv. Cat., _23, 209 (1973). 

5. M. L. Bender and M. Komiyama, " Cyclodextrin Chemistry," 
Springer-Verlag, Berlin, Heidelberg, New York, 1978. 

6. J. M. Lehn, Struct. Bonding (Berlin), _16, 1 (1973). 

7. J. M. Lehn, Accts. Chem. Res., 11, 49 (1978). 

8. J. M. Lehn and J. P. Sauvage, J. Am. Chem. Soc, 9_7, 67 00 (1975) 

9. E. Kauffmann, J. M. Lehn and J. P. Sauvage, Helv. Chim. Acta, 
5_9, 1099 (1976) . 

10. V. M. Loyola, R. Pizer and R. G. Wilkins, J. Am. Chem. Soc, 99 , 
7185 (1977). 

11. B. G. Cox, H. Schneidner and J. Stroka, J. Am. Chem. Soc, 100 , 
4746 (1978) . 

12. O. A. Gansow, A. R. Kauser, K. M. Triplet, M. J. Weaver, and 
E. L. Yee, J. Am. Chem. Soc, 9_9, 7087 (1977). 

13. F. J. Tehan, B. L. Barnett and J. L. Dve, J. Am. Chem. Soc, 
9_6, 7203 (1974) . 

14. J. L. Dye, M. R. Yemen, M. G. DaGue, and J. M. Lehn, J. Chem. 
Phys. , £8, 1665 (1978) . 

15. R. G. Teller, R. G. Finke, J. P. Collman, H. B. Chin and R. Bau, 
J. Am. Chem. Soc, 99^ H04 (1977). 

16. P. C. Manor and W. Saenger, J. Am. Chem. Soc, 9_6, 3630 (1974). 

17. R. J. Bergeron, M. A. Channing and K. A. McGovern, J. Am. Chem. 
Soc. , 100 , 2878 (1978) . 

18. K. Harata, Bull. Chem. Soc, Jpn. , 5_0, 1416 (1977). 

19. I . Tabushi, Y. Kiyosuke, T. Sugimoto and K. Yamamura, J. Am. 
Chem. Soc, 100 , 916 (1978). 

20. M. Mikolajczvk and J. Drabowicz, J. Am. Chem. Soc, 100 , 2510 
(1978). 



21. I. Tabushi, K. Fujita and H. Kawakubo, J. Am. Chem. Soc, 
99, 6456 (1977) . 

22. M. Komiyama and M.. Bender, J. Am. Chem. Soc, 9_9, 8021 (1977) 

23. R. Breslow, J. B. Doherty, G. Guillot and C. Lipsey, J. Am. 
Chem. Soc, 100 , 3227 (1978). 

24. Y. Matsui, T. Yokoi and K. Mochida, Chem. Lett., 1037 (1976). 



Intervalence Electron Transfer in i4ixed Valence Compounds 
Bruce C. Bunker Final Seminar October 19, 1978 

Experiments aimed, at understanding the energetics and rates 
of intervalence electron transfer in a variety of mixed-valence 
compounds have been undertaken to try to obtain a data base for 
testing current theories regarding more general electron transfer 
phenomena. These theories can be applied in areas of chemistry and 
physics ranging from electrochemistry to semiconductor science. 
In particular, an attempt has been made to check the validity and/or 
the limitations of the Hush theory , which relates the energy of 
the intervalence transfer band observed in the electronic spectrum 
to the energy of the thermal barrier to intervaj.ence electron trans- 
fer within the mixed- valence compound. 

Part of the research effort has been directed at studying 
mixed-valence compounds wnich show behavior which might be indi- 
cative of a breakdown in the Hush tneory. One such compound is the 
mixed-valence ^2,31 oxidation state of the dimer ^i-pyrazine-bis 
(pentaammine ruthenium) tosylate, wnich contains one Ru(Il) and one 
Ru(III). Since Qreutz and Taube first reported the synthesis of 
this dimer. in 1969, it has been the subject of a great deal of in- 
terest arid controversy in the literature. It exhibits an inter- 
valence transfer band which has neither the band width nor the sol- 
vent dependence predicted on the Dasis of tne Hush theory. The 
many attempts which have oeen reported to experimentally determine 
the rate of intervalence electron transfer have xed to ambiguous 
results. 

To further our understanding of this compound and the Hush 

tneory, we conducted experiments to help us formulate a description 

for the molecular orbitals involved in the electron transfer and to 

determine the rate of intervalence exectron transfer between the two 

metal centers. Variable temperature magnetic susceptibility results 

on the [2,3}pyr dimer coupled with low temperature EPK results for 

both tne [2,3] pyr and (3>3Jpyr dimers indicate that the odd electron 

in the dimer resides in the (d ,d ) orbital set on the Ru(ll) 

v xz' yz 

center and is not in a molecular oroital which is ueloealized over 

4 
both metals as postulated by Hush . Variable temperature studies 



indicate that in solution the rate of electron transfer from the 

Ru(II) center to the Ru(lII) center is fast on the NMR time scale 

•"5 o 

of 10 sec from room temperature uown to -80 C. It is slow on the 

-9 

EPR time scale of 10 sec in the solid state at all temperatures 

below -50 G. If outer coordination sphere effects are assumed to 
be negligible, then these EPR and iJMR results can be used to bracket 
the magnitude of the thermal energy barrier to between 3.4 and 6.7 
iccal/mole. This is in agreement with the Hush tneory prediction of 
4.5 kcai/mole, which is cased on the energy of the IT band. However, 
it may be that in the solid state, outer coordination sphere effects 
dominate, and that lattice effects can lock the odd electron onto 
one valence site. 

The other part of this research effort concerning mixed- 
valence compounds has been focused on trying to develop better ex- 
perimental techniques for studying tne intervaience electron trans- 
fer phenomenon. The technique which has been investigated most ex- 

5 6 
tensively is time domain rei'lectometry , or TDR . ' . This technique 

can be used to study the dielectric relaxation properties of sam- 
ples in the frequency range from 10 Hz up to 10 Hz. The charge 
oscillations associated with intervaience electron transfer can 
give rise to such dielectric relaxation. Analysis of TDR data can 
yield the relaxation frequency associated with the relaxation pro- 
cess, which can be used to calculate the rate of electron transfer. 
Variable temperature TDR data can be used to calculate the magni- 
tude of the thermal energy barrier to electron transfer by using 
the Arrhenius equation. 

TDR data has oeen obtained for a sample of europium sulfide 

which analyzes as Bu, S q . Analysis of tne data indicates that the 

9 
rate of electron transfer in this mixed valence compound is 2.2 x 10 

sec" at room temperature, and that the magnitude of the thermal 
energy barrier to the electron transfer process is around 920 cal/mole. 
This barrier corresponds almost exactly to the band gap energy 
associated with the semiconductivity of the compound, This has been 
determined from IR measurements, where the edge of the band gap ab- 
sorption feature can be observed at around 320 cm" . These results 
represent the first reported use of time uomain re flee tome try on 
solid f imples, and the first time the technique has been used to 



8 



determine the rate of electron transfer in a mixed valence compound. 

Other mixed-valence compounds have also been studied. Mag- 
netite, or Fe^O. , exhibits a rapid relaxation process which appears 

to be fast on the TDR time scale. This result is consistent with 

7 
Mossbauer results which nave been obtained for the compound • 

Soluble Prussian blue exhibits a relaxation process which is ob- 
servable in the TDR, but is slow on the TDR time scale of 10~ sec, 

Q 

which also agrees with the Mossbauer results . No relaxation is 
observed for solid U (l^H^) c -Hu)pPyr^ tosylate. This is consistent 
with the EPR study which indicates that the odd electron in the 
dimer is localized on one metal center in the solid state. The 
mixed-valence trimer Pe 2+ Fe 2 5+ 0(CH 3 COO)g(H ? K> reported by Gol 1 - 
danskii to exhibit dynamic electron transfer on the Mossbauer time 
scale, has also been studied. It does not exhibit the expected di- 
electric relaxation in the TDR, nor do several other mixed-valence 
oligamers which have been studied. It is currently not known whether 
This means that the TDR technique is only sensitive to electron trans- 
fer in extended lattice systems, or that literature estimates of 
electron transfer rates are incorrect for these systems. 

References 

1. N. Hush, Prog. Inorg. Chem. , 8, 591 (1967) 

2. C. Creutz and H. Taube, J. Am. Cnem. Soc, $1, 3988 (1969) 

3. B. Bunker, R. IS. Drago, D. N. Hendrickson, R. M. Richman, and 
3. L. Kessell, J. Am. Chem. Soc, 100, 3805 (1978) 

4. J. Beattie, N. Hush, and P. Taylor, Inorg. Chem., 1£, 992 (1976) 

5. M. J. C. van Gemert, Phillips Res. Repts., 28, 530 (1973) 

6. R. H. Cole, J. Phys. Chem., 22, 1459, 1469 (1975) 

7. R. Bauminger, et. al., Phys. Rev., 122, 1447 (1961) 

8. A. Ito, M. Suenaga, and K. Ono, J. Chem. Phys., 48, 3597 (1968) 

9. V. I. Czol ' danskii , et. al., Dokl. Akad. Nauk. S3SR (Phys. Chem.), 
211, 1063 (1973) 



PHOTO-OXIDATION OF INORGANIC COMPOUNDS 
WITH APPLICATION TO SOLAR ENERGY 

Richard W. Wegman October 26, 1978 

Thermal conversion and quantum conversion are two fundamental 
ways for converting solar energy into useful energy. A thermal sys- 
tem utilizes a collector that transforms light energy into heat. l 
The heat is transferred to a thermal storage system for use at a 
later time. Quantum conversion utilizes the available solar energy 
to induce a photochemical or photoelectric process in the absorbing 
material. In a photochemical process the absorber is activated to 
an excited state by the absorption of light. The energy gained by 
the absorber is lost by deactivation of the excited state through 
radiative and non-radiative decay processes. 2 An energy storage 
system coupled to the excited state provides an alternative deacti- 
vation pathway by directing the excess energy into a process that 
produces a stable fuel. 3 The efficiency of the energy storage process 
is limited by certain thermodynamic and kinetic parameters of the 
absorbing system. The thermodynamic limitations include: 

1) The maximum chemical potential difference achievable between 
the ground and excited state; 1 * and 

2) Power drainage into the storage system and its effect on 
the photo-generated chemical potential. 5 

The kinetic limitations arise from the inherent microreversibility of 
the system. 6 

The photo-dissociation of water into 2 and H 2 is a very attractive 
system for solar energy conversion. 7 Hydrogen is an excellent fuel 
and the source is cheap and plentiful. Unfortunately, high energy 
photons, not abundant in solar radiation, are required for direct 
photo-decomposition of water. Chemical systems that photo-catalytically 
decompose water by utilizing lower energy photons are therefore necessary 
One of the oldest known systems involves the photo-oxidation of Ce(III) 
to Ce(IV) in dilute perchloric acid solution. The process begins with 
the photo-excitation of Ce(III) to *Ce(III). The *Ce(III) ion reduces 
H + resulting in the production of H 2 gas and Ce(IV). The system is 
cyclic because Ce(IV) is capable of oxidizing water to regenerate H + 
and Ce(III) . 

A similar system involves the photo-oxidation of Cu(I) to Cu(II) 
in the presence of H + . 9 This process is non-cyclic because Cu(II) 
cannot oxidize water to regenerate starting materials. A mechanism 
for H 2 production in this system has recently been proposed. 10 

The photo-excitation of Ru(bpy) 2 . to *Ru (bpy) 3 and the conse- 
quent reduction of water by *Ru(bpy) 2+ has been suggested as a 
plausible method for water decomposition. 1 l Apparently, for kinetic 
reasons, the electron transfer process is too slow to compete with 
excited state decay. Recently, Meyer 12 has reported a working 



10 



2 + 



photo-electrochemical cell utilizing Ru(bpy)j . The cathode is a 
solution containing Ru(bpy)g + and Co(C 2 CU) 3 3 ~ in 1 N H 2 SO(t. The anode 
is a solution containing Fe(II) and Fe(III) 2 also in 1 N H2SO4. 
Photolysis_of the cathode produces *Ru(bpy)3 which is oxidized by 
Co(C 2 0it)3 3 to give Ru(bpy)3 + . A current is produced by the oxi- 
dation of Fe(II) by Ru(bpy)3 + . 

The photochemistry of di-nuclear metal complexes is of current 
interest and recently a rhodium 1 3 and several molybdenum di-nuclear 
compounds 14 ' 15 have been shown to stoichiometrically reduce H + to H 2 
during photolysis. 

The systems developed thus far have inherent problems which render 
them inadequate for the practical production of fuels. These problems 
include: 

1) High energy excitation wavelengths; 
• 2) Low quantum yields in the conversion process; 
3) Non-cyclic systems. 

Research oriented towards the engineering of systems capable of working 
at a practical level is necessary. 

References 

1. J. Richard Williams, Solar Energy, Technology and Applications , 
Ann Arbor Science Publishers, Inc., 1977. 

2. J. R. Bolton, J. Solid State Chem. , 2_2, 3 (1977). 

3. J. R. Bolton, to be published. 

4. R. T. Ross and Ta-Lee Hsiao, J. Appl. Phys., £8, 4783 (1977). 

5. R. T. Ross, J. Chem. Phys., £6, 4590 (1967). 

6. M. Almgren, Photochemistry and Photobiology , 21_, 603 (1978). 

7. V. Balzani, L. Moggi , M. F. Manfrin, F. Bolletta and M. Gleria, 
Science, 189 , 852 (1975) . 

8. L. J. Heidt and A. F. McMillan, J. Am. Chem. Soc, 7^6, 2135 (1954). 

9. D. D. Davis, G. K. King, K. L. Stevenson, E. R. Birnbaum, and 
J. H. Hageman, J. Solid State Chem., 22./ 63 (1977). 

10. G. Ferraudi, Inorg. Chem., £7, 1370 (1978). 

11. C. Creutz and N. Sutin, Proc. Nat. Acad. Sci., USA, Tl, 2858 (1975). 

12. B. Durham and T. J. Meyer, J. Am. Chem. Soc, 100 , 6286 (1978). 

13. K. R.' Mann, N. S. Lewis, V. M. Miskowski, D. K. Erwin, G. S. Hammond, 
and H. B. Gray, J. Am. Chem. Soc, 99, 5525 (1977). 



11 



14. D. K. Erwin, G. L. Geoffroy, H. B. Gray, G. S. Hammond, E. I. 
Solomon, W. C. Trogler and A. A. Zagars, ibid . , 99 , 3620 (1977). 

15. W. C. Trogler, D. K. Erwin, G. L. Geoffroy and H. B. Gray, ibid . , 
100, 1160 (1978) . 



12 

DYNAMICS OF ELECTRONIC SPIN TRANSITIONS 
IN IRON "SPIN-EQUILIBRIUM" SYSTEMS 

Wayne D. Federer November 9, 19 7 8 

Introduction 

Many iron (II) and iron(III) complexes exhibiting properties 
characteristic of the high- spin-- low- spin crossover region have 
been reported. 1 Most of these studies have focused on two questions: 
1) What types of ligands give rise to such behavior? and 2) How does 
variation of ligand substituents affect the position of the equilibrium? 
Only very recently has there been significant progress in answering a 
fundamental yet more difficult question, namely, what are the kinetics 
and mechanisms of spin state interconversions? Spin transitions are 
believed to play an important role in biological systems, and spin 
equilibria in certain hemoproteins may be coupled to electronic trans- 
port. 2 Essential to an assessment of the role of spin multiplicity 
changes in naturally occurring processes is an understanding of their 
dynamics in model systems. 

Solution State 

Spin transitions for isolated molecules may be viewed as inter- 
system crossing processes involving nonadiabatic electron transfer 
between two electronic isomers possessing distinct nuclear geometries. 3 
For systems in solution , magnetic data yield approximately linear log K 
vs. 1/T plots, consistent with a dynamic equilibrium. The "spin- 
flipping" rate has been measured directly for a number of complexes 
by the stimulated laser Raman T- jump technique. 3 These solution data 
generally indicate slower rates for those transitions involving greater 
changes in the metal-ligand bond lengths. Nonbonding intraligand 
steric interactions and multidentate chelation 1 * have been shown to 
retard the reorganization of the primary coordination sphere. In the 
absence of such stereochemical restraints, accelerated rates controlled 
by electronic factors are observed. Partial quantum-mechanical mixing 
of the states increases the rate of nonadiabatic electron transfer, with 
covalent bonding, especially to pi-acceptor sulfur ligands, leading to 
the fastest rates. Most recently, more accurate ultrasonic relaxation 
measurements at several temperatures have yielded estimates in the range 
10~ to 10~ 3 for the transmission coefficient for the spin-forbidden 
intersystem crossing in [Fe 11 (HB (pz) 3 )J , [Fe 11 (paptH) 2 ] + / and 
[Fe IH (Sal 2 trien) ] + . 5 / 5 

Solid State 

"Spin-equilibrium" compounds in the solid state seldom give linear 
plots of log K vs . 1/T and their physical properties cannot be readily 
interpreted in terms of independently acting molecules. Thermally in- 
complete transitions due to "residual paramagnetic impurities" , lb 
"plateaus" due to "lattice effects of unknown origin" , H abrupt phase 
transitions, 7 thermal hysteresis, 8 and marked dependence of the position 
and nature of the "equilibrium" on counter ion 1 * 3 and solvation effects 9 
have intrigued and bewildered many workers. 



13 



Several lines of evidence suggest that spin transitions in the 
solid state are cooperative in nature, with significant coupling 
between the electronic states and lattice vibrations: 

1) In [Fe 111 (phen) 2 (NCS) 2 ] / which undergoes an almost dis- 
continuous spin transition, heat capacity measurements reveal a X-type 
phase transition accompanied by an entropy change far greater than can 
be accounted for by magnetic and conf igurational entropy alone. Infra- 
red spectra throughout the transition region show two coexisting spin 
states, each characterized by a constant 10 Dq value. 7 

2) In [Fe 11 (4-CH 3 -phen) 2 (NCS) 2 ] / which exhibits a continuous 
transition over a 125° range, the Debye-Waller factor for each spin 
state shows deviations from the Debye model, but only when the site 
fraction of that particular state drops below a limiting value. 10 ' 11 
The nature of these deviations is taken as evidence for the formation 
of domains of the minority spin state. Further evidence for the 

presence of domains is provided by an ESCA study of [Fe 11 ( 6-MePy) 3 tren] PF 6 . 12 

3) Mossbauer studies 13 show that the transition temperature, 
T c , for the gradual spin multiplicity change in mixed crystals of 

[Fe x Zn]__ x ( 2-pic) 3 ]Cl2 'C2H50H decreases with dilution of the ferrous ion, 
consistent with a smaller total enthalpy change for lower concentrations. 

4) The pronounced hysteresis (AT C = 34°) seen for 
[Fe(paptH) 2 ] (NC>3)2 has made possible a unique Mossbauer kinetic study 

of the high-spin to low-spin transition rate. 8 Extrapolation of the 
data to 298K yields a rate 2 x 10 5 times slower than that observed 
in solution. Extrapolation to 105K yields a spin lifetime of 19 years! 
These results show that thermally incomplete transitions are merely 
kinetically controlled non- equilibrium systems. 

The observations are all consistent with a static mechanism for 
the solid state. Occasional thermal excitations induce localized 
molecular distortions which are "communicated" throughout the lattice 
via long-wave phonons to form domains which may eventually combine. 
The strength of the coupling determines the sharpness of the transition. 



10/ 11 



Fundamental unanswered questions concern the mechanisms of spin 
interconversion for 1) the extensively studied ferric dithiocarbamates ! 4 
and a few other systems 15 which appear to be dynamic equilibria in 
solids, and 2) those rare systems which show continuously varying 10 Dq 
over the transition region. Also of interest is the possibility of 
coupling of the phonons of a dense medium 17 ' 18 with the "dynamic" spin 
transitions in solutions . Finally, the relevance of these studies of 
spin transition dynamics to biological systems is briefly considered. 19-21 

References 

1. For the most recent reviews, see a) R. H. Martin and A. H. White, 
"Transition Metal Chemistry", 4_, 113 (1968); b) H. A. Goodwin, 
Coord. Chem. Rev., 18, 293 (1976). 



14 



2. E. V. Dose, M. F. Tweedle, L. J. Wilson, and N. Sutin, J. Am. 
Chem. Soc, 99^, 3887 (1977). 

3. E. V. Dose, M. A. Hoselton, N. Sutin, M. F. Tweedle, and L. J. 
Wilson, J. Am. Chem. Soc, 100 , 1141 (1978), and references 
therein. 

4. R. H. Petty, E. V. Dose, M. F. Tweedle, and L. J. Wilson, Inorg. 
Chem., 17, 1064 (1978). 

5. J. K. Beattie, R. A. Binstead, and R. J. West, J. Am. Chem. Soc, 
100 , 3044 (1978) . 

6. R. A. Binstead, J. K. Beattie, E. V. Dose, M. F. Tweedle, and 
L. J. Wilson, J. Am. Chem. Soc, 190 , 5609 (1978). 

7. M. Sorai and S. Seki, J. Phys . Chem. Solids, 35, 555 (1974) . 

8. G. Ritter, E. Konig, W. Irler, and H. A. Goodwin, Inorg. Chem., 
17 , 224 (1978) . 

9. R. J. Butcher and E. Sinn, J. Am. Chem. Soc, 9_8, 5159 (1976). 

10. B. Kanellakopulos , E. Konig, G. Ritter and W. Irler, Journal de 
Physique (Suppl.), 21_, C6-475 (1976). 

11. E. Konig, G. Ritter, W. Irler, and B. Kanellakopulos, J. Phys. C: 
Solid State Phys., 10, 603 (1977). 

12. M. A. Hoselton, R. S. Drago, L. J. Wilson, and N. Sutin, J. Am. 
Chem. Soc, 9J_, 6967 (1976). 

13. P. Gutlich, R. Link, and H. G. Steinhauser, Inorg. Chem., 17 , 
2509 (1978) . 

14. M. Sorai, J. Inorg. Nucl. Chem., 4_0, 1031 (1978), and references 
therein. 

15. See, for example, M. Eibschutz and F. J. Disalvo, Phys. Rev. Lett., 
36, 104 (1976) . 

16. R. Morassi and L. Sacconi, J. Am. Chem. Soc, 9_2, 5241 (1970). 

17. J. Jortner, J. Chem. Phys., 6_4, 4860 (1976). 

18. A. Raap, J. W. van Leeuwen, H. S. Rollema, and S. H. DeBruin, FEBS 
Lett. , 81, 111 (1977) . 

19. P. M. Champion, E. Munch, P. G. Debrunner, P. F. Hollenberg, and 
L. P. Hager, Biochemistry, 12, 426 (1973) . 

20. J. P. Collman, T. N. Sorrell, K. 0. Hodgson, A. K. Kulshrestha, and 
C. E. Strouse, J. Am. Chem. Soc, 9j), 5180 (1977). 

21. M. F. Perutz, et al., Biochemistry, 17, 3640, 3652 (1978). 



AQUEOUS MANGANESE (III) COMPLEXES WITH OXYGEN DONOR LIGANDS 
Brenda R. Shaw Suib November 16, 19 78 

Because the hexaaquo manganese (III) ion is very unstable 
toward disproportionation 

2Mn(III) + 2H 2 t Mn(II) + MnO 2 + 4H + (1) 

and reduction (E° *1.5 volts) in water at neutral pH , aqueous 
Mn(III) species are not well known. Recently, however, the im- 
portance of Mn(III) complexes in biological and environmental 
systems has been postulated and in some cases verified. For 
these reasons, inorganic chemists have recently begun looking in 
earnest for stable manganese (III) complexes in order to explore 
the chemistry of these little-known species. Because complexes 
of Mn(III) with oxygen donor ligands appear to be the most impor- 
tant in biological systems and because" they have been the most 
widely studied, the present discussion wi"ll be limited to these 
species . 

Three recent reviews 1 ' 2 ' 3 describe how manganese (III) can be 
stabilized with respect to disproportionation and reduction in 
aqueous solution. Disproportionation can be prevented by the use 
of H + and Mn(II) to shift the equilibrium (see equation 1) toward 
the left or by complexing the manganese (III ) with non-oxidizable 
ligands to prevent autoreduction. The Mn (III) /Mn (II) couple is a 
measure of the stability of the Mn(III) complexes and some ligands 
can lower this reduction potential by as much as 0.7 volt. 1 This 
decrease in reduction potential allows manganese (III ) complexes 
to exist in aqueous solution. 

Besides thermodynamically stabilizing Mn(III) toward reduction, 
certain ligands lower the kinetic lability of Mn(III). 4 This effect 
is demonstrated by a kinetic study in which pyrophosphate competes 
effectively with the oxidizable substrate iso-propyl mandelate for 
a binding site on Mn(III) . 5 The iso-propyl mandelate must bind to 
the Mn(III) in order to be oxidized and released as benzaldehyde . 
Work by Jones and Hamm" and Pelizzetti, et. al. 6 shows that the re- 
duction of manganese (III) is inhibited by H^" ions and chelating 
ligands which is often observed for trivalent transition metal ions. 
The H + ion effect is due to the equilibrium shown below: 



Mn(III) + H 2 t MnOH 2+ + H + (2) 



The MnOH 2 is often a more reactive species toward reduction than 
hexaaquo Mn(III) so the forward reaction will be enhanced by higher 
pH . Strong complexing species affect the lability of Mn(III) toward 
reduction as shown in the following example: 

[Mn(III) (substrate) ] t Mn(II) + oxidized substrate (3) 

When a strong chelating ligand is present, the reverse reaction be- 
comes more favorable with respect to the forward reaction since the 
'more highly charged Mn(III) can form stronger complexes than Mn(II). 



16 

From these examples, one sees that factors affecting thermodynamic 
stability of Mn(III) toward reduction may be the same factors that 
affect kinetic lability of these complexes. 

The manganese (III) gluconate complex prepared by Sawyer and 
coworkers is both kinetically and thermodyamically stable toward 
oxidation and reduction even in the presence of hydrogen peroxide. 7 
This complex exists in a monomer-dimer equilibrium as shown by 
polarography , absorption spectra and magnetic susceptibility 
measurements. 8 ' 9 In the monomer form it is stable in the presence of 
oxygen, while the binuclear species is oxidized to give 
the manganese (IV) complex. The monomeric gluconate 
complex is the most stable toward reduction of the many complexes 
of manganese (III) prepared by D. T. Sawyer and coworkers. This 
group has also characterized a large group of complexes with other 
polyhydroxy ligands using polarography and magnetic susceptibility 
experiments. 10 The relevance of their work to plant photosynthesis 
has been summarized along with recommendations for obtaining a 
reasonable inorganic model for photosystem II. 11 

Photosystem II is the biological system receiving the most 
attention with respect to manganese (III) . Proton relaxation 
studies on this system show that the manganese atoms in the Mn- 
enzyme of photosystem II exist in several oxidation states both 
in the resting and in the active enzyme. 12 While manganese (III ) 
is likely one of the oxidation states present in photosystem II, 
more conclusive evidence exists for its presence in two other biological 
systems. The enzyme manganosuperoxide dismutase was shown to con- 
tain Mn(III) in the resting state by epr and magnetic measurements 
on the purified enzyme. 13 Resonance Raman spectra show that 
manganese (III) can be substituted for Fe(III) in ovo- and human 
serum transferrins . l u 

Manganese (III) complexes also appear to play an important 
role 15 ' 16 in the complicated chemistry of manganese in soil and 
surface waters 17 ' 18 ' 19 although further work is needed. Since Mn(II) 
and Fe(III) form stable outer sphere and inner sphere complexes 
with fulvic acids respectively, 20 it is reasonable that Mn(III) 
might also be stabilized by fulvic acids in natural waters. This possi- 
bility could be tested using water proton nmr line widths in studies 
similar to those reported by Gamble, Langford and Tong." 



2 



References 

1. G. Davies, Some aspects of the chemistry of manganese (III) in 
aqueous solution, Coordin. Chem. Rev ., 4_, 199 (1968) . 

2. A. McCauley, The role of complexes in metal-ion oxidations 
in solution, Coordin. Chem. Rev ., 5^, 245 (1970). 

3. W. Levason and C. A. McAuliffe, Higher oxidation state chemistry 
of manganese, Coord. Chem. Rev ., 1_, 353 (1972). 



17 

4. T. E. Jones and R. E. Hamm, Kinetics of the reaction between 
1, 2-diaminocyclohexanetetraacetatomanganate (III) ion and 
hydrogen peroxide, Inorg . Chem . , 13 , 1940 (1974). 

5. R. D. Malkani, K. S. Suresh and G. V. Bakore, Kinetics and 
mechanism of oxidation of iso-propyl mandelate by manganese (III) 
pyrophsophate, J. Inorg. Nucl. Chem. , 39 , 621 (1977) . 

6. E. Pelizzetti, E. Mentasti and E. Pramauro, Kinetics and 
mechanism of oxidation of ascorbic acid by manganese (III) 

in aqueous acidic perchlorate media, J.C.S. Dalton , 61 (1978) . 

7. M. E. Bodini and D. T. Sawyer, Electrochemical and spectro- 
scopic studies of manganese (II) , -(III), and -(IV) gluconate, 
2. Reactivity and equilibria with molecular oxygen and 
hydrogen peroxide, J. Amer. Chem.-Soc , 98 , 8366 (1976) . 

8. D. T. Sawyer and M. E. Bodini, Manganese (II) gluconate. Redox 
model for photosynthetic oxygen evolution, J. Amer. Chem. Soc , 
97, 6588 (1975) . 

9. M. E. Bodini, L. A. Willis, T. L. Riechel and D. T. Sawyer, 
Electrochemical and spectroscopic studies of manganese (II) , 
-(III), and -(IV) gluconate complexes. 1. Formulas and 
oxidation-reduction stoichiometry , Inorg. Chem . , 15 , 1538 
(1976) . 

10. K. D. Magers, C. G. Smith and D. T. Sawyer, Polarographic 
and spectroscopic studies of the manganese (II ) , -(III), and 
-(IV) complexes formed by polyhydroxy ligands, Inorg . Chem . , 
17, 515 (1978) . 

11. D. T. Sawyer, M. E. Bodini, L. A. Willis, T. L. Riechel and 
K. D. Magers, Electrochemical and spectroscopic studies of 
manganese (II , III , IV) complexes as models for the photosynthetic 
oxygen-evolution reaction, in Adv. Chem. Ser. , no. 162 , 
Bioinorganic Chemistry, Ed. K. N. Raymond, A.C.S., Washington, 
D.C., 1977, pp. 330-349. 

12. T. Wydrzynski, N. Zumbulyadis, P. G. Schmidt, H. S. Gutowski, 
and Govindjee, Proton relaxation and charge accumulation during 
oxygen evolution in photosynthesis, Proc. Natl. Acad. Sci. , 
USA , 73, 1196 (1976) . 

13. J. A. Fee, E. R. Shapiro, and T. H. Moss, Direct evidence for 
manganese (III) binding to the manganosuperoxide dismutase of 
Escherichia coli B, J. Biol. Chem . , 251 , 6157 (1976). 

14. Y. Tomimatsu, S. Kint and J. R. Scherer, Resonance Raman 
spectra of iron(III)-, copper(II)-, cobalt (III) - , and 
manganese (III ) -transferrins and of bis (2,4,6- 
trichlorophenolato) diimidazolecopper (II) monohydrate, a 
possible model for copper (II) binding to transferrins, 
Biochemistry , 15 , 4918 (1976). 



18 

15. W. Stumm and J. J. Morgan, Aquatic Chemistry , Wiley-Interscience, 
New York, 1970, p. 525. 

16. L. H. P. Jones and G. W. Leeper, The availability of various 
manganese oxides to plants, Plant and Soil , 3_, 141 (1951) . 

17. H. Bilinski and J. J. Morgan, Complex formation and oxygenation 
of manganese (II) , Presented before the Division of Water, Air, 
and Waste Chemistry, American Chemical Society, Minneapolis, 
April, 1969. 

18. S. M. Bromfield, The properties of a biologically formed 
manganese oxide, its availability to oats and its solution 
by root washings, Plant and Soil , 9_, 325 (1958) . 

19. S. M. Bromfield and D. J. David, Sorption and oxidation of 
manganous ions and reduction of manganese oxide by cell 
suspensions of a manganese oxidizing 'bacterium, Soil Biol . 
Biochem . , 8_, 37 (1976). 

20. D. S. Gamble, C. H. Langford, and J. P. K. Tong, The structure 
and equilibria of a manganese (II) complex of fulvic acid studied 
by ion exchange and nuclear magnetic resonance, Can. J. Chem . , 
54, 1239 (1976) . 



19 

THERMOCHROMIC BEHAVIOR OF SELECTED INORGANIC SYSTEMS 
David M. Hamilton, Jr. November 28, 1978 



The relationship between the color and structure of inorganic 
coordination compounds has generated interest in the study of the 
phenomenon known as thermochromism. Thermochromism has been defined 
as the reversible change in the color of a compound which has been 
either heated or cooled. 1 ' 2 Observations have shown that this color 
change may occur either gradually, characteristic of continuous 
thermochromism, or abruptly, characteristic of discontinuous thermo- 
chromism. More specifically, the thermochromic behavior of inorganic 
systems has been attributed to either (1) a change in the ligand 
geometry of a coordination complex, (2) a temperature dependence of 
the line widths of electronic absorption bands, (3) a variation in 
the number of solvent molecules in the coordination sphere of a sol- 
vated metal ion or (4) a shift in an equilibrium between two different 
molecular structures. 2 

Much of the recent research in the area of thermochromic inorganic 
compounds has been centered around the N ,N-diethylethylenediamine 
complexes of copper (II) 3 ~ 7 and nickel (II) 3 ' 8 salts. Pfeiffer and 
Glaser, 9 in the course of their preparative work with copper salts 
of complex nitrogen ligands , discovered the copper (II) perchlorate 
salt of the bidentate ligand, N,N-diethylethylenediamine (dieten, 
H 2 NCH 2 CH 2 N (C 2 H 5 ) 2 ) , was thermochromic. The early work of these 
investigators was later modified by Lever, et al. , 4 who succeeded 
in preparing two additional thermochromic compounds, Cu (dieten) 2 (BF 4 ) 2 
and Cu (dieten) 2 (NO 3 ) 2 . The observed color change for each of these 
compounds was from red to violet as the temperature was increased. 
Analogous Ni (dieten) 2 X 2 complexes have also been prepared, 3 ' 8 where 
X = I, C10 U , BF U/ NO 3 , Br, Agl 2 , Pbl 3 , ^Hgl u , and ^CdBru . All of the 
nickel complexes were found to be thermochromic changing color from 
orange-yellow to bright red at elevated temperatures. 

The nature of the thermochromic behavior of both the copper 
and nickel compounds was investigated using a variety of techniques, 
including: infrared spectroscopy, 4 ' differential scanning calorimetry , 3 
EPR spectroscopy 5 and magnetic susceptibility. 1 * It was determined 
from these experiments that the thermochromic transitions resulted 
from increased axial interaction between the metal-nitrogen cationic 
species and the anionic species of the given complexes. 3-6 It was 
noted that this increase in axial perturbation correlated well with 
an observed decrease in the infrared stretching frequencies of the 
metal-nitrogen bonds. 4 In addition, differential scanning calorimetric 
data confirmed the observations of discontinuous thermochromism for 
the complexes, Cu(dieten) 2 X 2 , X = BF u , C 10 u and NO 3 and Ni (dieten) 2 X 2 , 
X = BF U and C10 U . The other compounds, with the exception of 
Ni (dieten) 2 (N0 3 ) 2 which decomposed before its transition temperature 
was reached, were continuously thermochromic. These results were 
explained in terms of anion polarizabili ty . 3 



20 



The partially substituted ammonium salts of the tetrachloro- 
cuprate 10-15 and tetrachloronickelate x 6 anions have also received 
attention because of their thermochromic behavior. The copper 
salts studied were prepared by the method of Remy and Laves, 17 
who obtained crystalline compounds by mixing CuCl 2 *2H 2 with the 
appropriate amine hydrochloride in alcoholic solution. From a 
number of X-ray studies 1 x ' 1 3 ' l 8 ~ 2 x on these copper complexes, it 
was determined that the symmetry of the tetrachlorocuprate anion 
ranged from tetrahedral 22 to square planar. 13 However, the most 
stable of the possible symmetries for the CuCl 2 anion was D2d. 18 22 
The change in the color of most of these compounds was directly 
related to the degree of distortion of the CuClu anion away from 
D2d symmetry. This fact led to attempts at correlating the amount 
of anion distortion with observed electronic absorption spectra. 10 ' 12 
Also, a pressure study was made to determine if these distortions 
could be induced by another method 14 During the course of one 
study, 10 it was found that the compound, (dimethylammonium) 3 CuCl 5 , 
although thermochromic, did not undergo any type of structural dis- 
tortion, but rather was thermochromic because the electronic band 
widths were temperature dependent. 

The nickel ammonium salts have been prepared only very recently 
from high temperature melts produced in sealed tube reactions. 16 
Complete single crystal X-ray data were not obtained for the nickel 
compounds, but from the preliminary powder data results the low 
temperature form was believed to be a distorted octahedral polymer 
with bridging halogens, while the high temperature form was thought 
to contain discrete NiCl 2- anions with tetrahedral geometry. It 
was also found that the nature of the cation in these complexes and 
its ability to interact with the anion, particularly the ability to 
hydrogen bond and whether the amine was partially substituted with 
alkyl or aryl groups, could alter the thermochromic characteristics 
of the complexes. 16 Here, the change in the color of the compounds 
resulted from the change in the ligand geometry of the NiCl 2 anion 
as the temperature was varied. Discontinuous thermochromism was 
apparent for all of these ammonium nickel salts. 

Lastly, thermochromic behavior has been observed for solutions 
of anhydrous CoCl 2 in water, methanol, 1-propanol, 2-propanol and 
acetone. 23 Dramatic changes in the colors of these solutions occurred 
when the temperature was lowered. These effects were a result of a 
change in the configuration of the solvated cobalt complexes from a 
high temperature tetrahedral form to a low temperature octahedral 
form. 

References 

(1) J. H. Day, Chem. Rev., £3, 65 (1963). 

(2) J. H. Day, Chem. Rev., £8, 649 (1968). 

(3) L. Fabbrizzi, M. Micheloni, and P. Paoletti, Inorg. Chem., 13 , 
3019 (1974) . 



21 



A. B. P. Lever, E. Mantovani, and J. C. Donini, Inorg. Chem. , 
10, 2424 (1971) . 

H. Yokoi, M. Sai, and T. Isobe, Bull. Chem. Soc. Jap., 42 , 
2232 (1969) . 

A. B. P. Lever and E. Mantovani, Inorg. Chem., 10_, 817 (1971). 

W. E. Hatfield, T. S. Piper, and V. Klabunde, Inorg. Chem., 2, 
629 (1963). 

D. M. L. Goodgame and L. Venanzi, J. Chem. Soc, 616 (1963) . 

P. Pfeiffer and H. Glaser, J. Prakt. Chem., 151 , 134 (1938). 

R. D. Willett, J. A. Haugen, J. Lebsock, and J. Morrey, Inorg. 
Chem. , 13, 2510 (1974) . 

D. N. Anderson and R. P. Willett, Inorg. Chim. Acta, S, 167 (1974). 

R. L. Harlow, W. L. Wells, III, G. W. Watt, and S. H. Simonsen, 
Inorg. Chem., 14, 1768 (1975). 

R. L. Harlow, W. L. Wells, III, G. W. Watt, and S. H. Simonsen, 
Inorg. Chem., 13_, 2106 (1974). 

R. D. Willett, J. R. Ferraro, and M. Choca, Inorg. Chem., 13 , 
2919 (1974). 

R. D. Willett, 0. L. Liles, Jr., and C. Michelson, Inorg. Chem., 
6, 1885 (1967) . 

J. R. Ferraro and A. T. Sherren, Inorg. Chem., 1/7, 2498 (1978). 

H. Remy and G. Laves, Ber. Deut. Chem., Ges . 3, 6_6, 401 (1933). 

R. D. Willett and M. L. Laresen, Inorg. Chim. Acta, 5_, 175 (1971) . 

J. Lamotte-Brasseur , L. Dupont, and O. Dideberg, Acta Crystallgr., 
Sect. B, 2_9, 241 (1973) . 

J. P. Steadman and R. D. Willett, Inorg. Chim. Acta, _4, 367 (1970). 

G. B. Berrell and B. Zaslow, J. Inorg. Nucl. Chem., 3J_' 1751 (1972) 

R. M. Clay, P. Murray-Rust, and J. Murray-Rust, J. Chem. Soc, 
Dalton Trans., 595 (1973). 

(23) W. C. Nieuwpoort, G. A. Wesselink, and E. H. A. M. Van der Wee, 
Rec Trav. Chem., 85, 397 (1966). 



(4) 


(5) 


(6) 


(7) 


(8) 


(9) 


(10) 


(11) 


(12) 


(13) 


(14) 


(15) 


(16) 


(17) 


(18) 


(19) 


(20) 


(21) 


(22) 



22 



PHYSICAL PROPERTIES OF THE ONE-DIMENSIONAL CONDUCTING MATERIAL 
BaVS 3 : ITINERANCY AND STOICHIOMETRY 

Jeff Kelber November 30, 1978 



"One-dimensional" (1-d) materials may be defined as materials in 
which the magnetic and electronic interactions are much larger along 
one crystallographic axis than the other two. Such materials are of 
theoretical interest 1 as physical approximations of 1-d magnetic 
lattices, and of technological interest as possible prototypes of 
high temperature superconductors. 2 ' 3 

The 1-d materials most studied to date 3 - TTF-TCNQ and derivatives, 
and the partially-oxidized Krogmann's salts - manifest a lattice 
instability, termed a Peierls distortion, 3 ' ^ thought to be inherent in 
all 1-d partially-filled band systems. This instability results in the 
formation of a superlattice along the unique crystallographic axis, and 
the opening of a band gap at the fermi surface, destroying the metallic 
state. Much research is now being directed towards synthesizing and 
characterizing new 1-d metals in order to learn more about the physics 
of the Peierls distortion and other 1-d cooperative phenomena. 

The room temperature structure 5 of B aVS 3 is hexagonal' (P63/mmc), 
with chains of face-sharing VS 6 octahedra running parallel to the c-axis 
and barium atoms separating the chains. BaVS 3 undergoes a distortion 
to orthohombic symmetry 5 at lower temperature. The intrachain V-V 
distance 5 (2.805 A) is short enough that one expects metallic interaction 
along the vanadium chains. 5 BaVS 3 is structurally quite different from 
TTF-TNQ and the Krogmann's salts, which consist of planar complexes 
stacked perpendicular to the unique axis. BaVS 3 is composed of chains 
of metal atoms bridged by sulfur ligands and may therefore be expected 
to reveal new aspects of 1-d cooperative phenomena. 

BaVS 3 can be synthesized by either of the following methods: 5 ' 7 

H 2 S 

(1) BaC0 3 + V 2 5 — £-» BaVS 3 (powder) 

(2) BaS + V + 2S Va ^ UUm > BaVS 3 (powder) 

Ferromagnetic BaVS 3 is characterized 7 by a Curie temperature near 
16K, and a magnetic moment in the ordered phase of ~0.2 Bohr magnetons 
(y ) /vanadium. Annealing this material in the presence of sulfur 7 
gradually alters the material to structurally similar antif erromagnetic 
BaVS 3 . Further annealing with sulfur produces no further change in the 
sample or its magnetic properties. (The process can be reversed by 
annealing in the absence of sulfur) . The antif erromagnetic state can 
also be synthesized directly by: (1) using long reaction times 8 or by 
(2) with an excess of sulfur, and analysis indicates that the anti- 
ferromagnetic material is stoichiometric BaVS 3 . 7 ' 8 



23 

Nonstoichiometric BaVS 3 7 is also characterized by anomolies in 
the resistivity near 16K, 70K, and 150K, and a lattice distortion 
temperature of ~150K. Stoichiometric BaVS 3 is characterized 7 ' 8 by 
an electronic transition and a broad maximum in the magnetic suscep- 
tibility near 70K and a lattice distortion temperature of -250K. An 
ESR signal (g -1.90) is observed only in nonstoichiometric BaVS 3 . 7 

Recent neutron diffraction measurements confirm that: 

(1) stoichiometric and nonstoichiometric BaVS 3 are structrually 
distinct (though similar); 

(2) the low moment observed in the nonstoichiometric material is 
not due to f errimagnetism; 

(3) No long range magnetic order occurs in stoichiometric BaVS 3 
above 8K. 

An itinerant electron model, with localized electrons trapped at 
lattice defects coupling through the itinerant vanadium 3-d electrons 
(a modified RKKY interaction 9 ) qualitatively explains the experimental 
data. 

The possibility of itinerancy in the BaVS 3 system is exciting 
because: (a) the lattice distortion is not coupled to the major electronic 
transition (near ~70K) , and so would not seem to be the Peierls 
distortion expected for a d 1 system with (formally) one unpaired electron 
per vanadium site; and (b) all other known metal halides and oxides 
with the BaVS 3 structure are insulators. 8 ' 10 

References 

1. L. deJong and A. R. Midiema, Experiments on Simple Magnetic Model 
Systems , Taylor and Franic, Ltd., London, 1977. 

2. W. A. Little, Phys . Rev., 134 , A1416 (1964). 

3. H. Keller (Ed.), Low Dimensional Cooperative Phenomena (NATO Advanced 
Studies Institute), Plenum Press, New York, 1975. 

4. R. E. Peierls, Quantum Theory of Solids , Oxford University Press, 
London, 1955, pp. 108-114. 

5. R. Gardner, M. Vlasse and A. Wold, Acta. Cryst. B25 , 781 (1969) . 

6. J. B. Goodenough, Magnetism and the Chemical Bond , Wiley Interscience, 
New York, 1963, pp. 249-295. 

7. 0. Massenet, R. Buder, J. J. Since, C. Schlenker, J. Mercier and 
J. Kelber and G. D. Stucky, Mater. Res. Bull., 13, 187 (1978). 

\ 8. M. Takano, H. Kosugi , N. Nakanishi, M. Shimada, T. Wada and 
M. Kiozumi, J. Phys. Soc. Japan, 43, 1101 (1977) . 



24 



9. A similar model has been proposed by A. Mauger and P. Hugon, 
Physica, 86-88B , 1007 (1977). 

10. M. Shimada, F. Kanamarv, and M. Kiozumi, Sol. St. Comm. , 18 , 
1561 (1976) . 



25 



SOME RECENT DEVELOPMENTS IN THE INTERCALATION CHEMISTRY 
OF THE LAYERED TRANSITION METAL DICHALCOGENIDES 



Deborah E. Samkoff December 7 , 1978 



Interest in the layered transition metal dichalcogenides originally 
derived from their highly anisotropic transport properties l . There 
are more than forty such compounds known 2 , chiefly (although not ex- 
clusively) dichalcogenides of the early transition metals. 

As in the case with other layered materials such as graphite 3 

and sheet silicates u , certain atoms and molecules may be intercalated 

into the interlamellar spaces of Group IV B and Group V B dichalco- 
genides 2 . 

Intercalation complexes of alkali and transition metal atoms in 
Group IV B VB disulfides and diselenides are of current interest as 
possible electrode materials in high energy density batteries 5 . 
Host-guest bonding in these complexes is generally thought 6 to pro- 
ceed with nearly complete electron transfer from guest to host. 

Molecular species which form intercalation complexes with 
layered transition metal disulfides and diselenides are mostly 
nitrogenous Lewis bases 7 . The bonding in these complexes was 
postulated from the first to involve "electron donation" from 
guest to host, but in the usual covalent sense of coordination 
complex formation 1 . Recent evidence for redox and ionization 
processes occurring in these complexes appears to apply only to 
ammonia 6 . Attempts to elucidate the details of bonding led to 
numerous structural studies, in which such techniques as powder 9 
and single crystal 10 ' 11 x-ray diffraction, single crystal neutron 
diffraction 12 , wide-line NMR 1 3 , and incoherent inelastic neutron 
scattering 1 have been employed to reveal patterns of guest orien- 
tation in the interlamellar spaces of the host and to piece together 
a bonding scheme which involves lone pair donation into the unfilled 
chalcogen valence band and which suggests that interactions among 
guest molecules can be important. 



Dines 15 has recently advanced the argument that host-to-guest 
ir-backbonding is also important in the formation of intercalation 
complexes by preparing and characterizing alkyl isonitrile inter- 
calates of TiS 2 and TaS 2 . In this connection, some early work in 
the field with phosphines 7 bears re-examination and, perhaps, 
repetition and extension. In any case, the bonding in the "molecular" 
intercalation complexes, even for such a simple intercalate as 
ammonia, cannot be considered a solved puzzle. 

Another class of compounds recently added to the list of those 
which will intercalate into layered transition metal dichalcogenides 
is the low ionization potential organometallic sandwich compounds-- 



26 



metallocenes x 6 ~ 1 8 , bis (arene) metal compounds 17 ' 18 , and some mixed 
sandwich compounds 17 . The sandwich compounds are thought to act 
as pseudo-alkali metals in intercalation. The organometallic 
sandwich compound intercalates offer another opportunity and 
avenue of approach to gauge the relative importance of guest-guest 
and guest-host interactions to the structure and bonding in 
layered transition metal dichalcogenide intercalation complexes 
and the possibility of reagents with usefully modified reactivity. 

The recent report by Chianelli and Dines 19 of low- temperature 
preparations of Group IV B, V B, and VI B dichalcogenides by 
solution metathesis reactions offers a route to several forms of 
layered dichalcogenides not available by the traditional high- 
temperature elemental syntheses x . The compounds formed in these 
low- temperature solution syntheses show reactivities, particularly 
toward intercalation, not previously accessible. 

References 

1 F . R. Gamble, F. J. DiSalvo, R. A. Klemm, and T. H. Geballe, 
Science, 168 , 568 (1970) . 

2 J. S. Wilson and A. D. Yoffe, Adv. Phys., _18, 193 (1969). 

3 A. R. Ubbelohde and F. A. Lewis, "Graphite and its Crystal 
Compounds," Clarendon Press, Oxford, England, 1960. 

U J. M. Thomas, J. M. Adams, S. H. Graham, and D. T. Tennako, Adv. 
Chem. Ser., 163 , 298 (1977). 

5 M. S. Whittingham, Science, 19 2 , 1126 (1976). 

6 B. G. Silbernagel, Solid State Commun. , r7, 361 (1975). 

7 F. R. Gamble, J. H. Osiecki, M. Cais, R. Pisharody, F. J. DiSalvo, 
and T. H. Geballe, Science, 174 , 493 (1971) . 

8 R. Schollhorn, and H. D. Zagefka, Angew. Chem. Int. Ed. Engl., 16 , 
199 (1977) . 

9 F. R. Gamble, J. H. Osiecki, and F. J. DiSalvo, J. Chem. Phys., 55 , 
3525 (1971) . 



1 



G. S. Parry, C. B. Scruby, and P. M. Williams, Philos. Mag., 29 , 
601 (1974) . 

1 R. R. Chianelli, J. C. Scanlon, M. S. Whittingham, and F. R. 
Gamble, Inorg. Chem., 14, 1691 (1975). 



27 



12 C. Riekel, D. Hohlwein, and R. Schollhorn, J. Chem. Soc. Chem. 
Commun. , 863 (1976) . 

13 B. G. Silbernagel, M. B. Dines, F. R. Gamble, L. A. Gebhard, and 
M. S. Whittingham, J. Chem. Phys . , 6j>, 1906 (1976); B. G. 
Silbernagel and F. R. Gamble, ibid . , 1014. 

1U B. C. Tofield and C. J. Wright, Solid state Commun., 22_, 715 (1977) 

15 M. B. Dines, Inorg. Chem., 17, 762 (1978). 

16 M. B. Dines, Science, 188, 1210 (1975). 

17 W. B. Davies, M. L. H. Green, and A. J. Jacobson, J. Chem. Soc. 
Chem. Commun., 781 (1976). 

18 Ri P. Clement, W. B. Davies, K. A. Ford, M. L. H. Green, and 
A. J. Jacobson, Inorg. Chem., r7, 2754 (1978). 

19 R. R. Chianelli and M. B. Dines, Inorg. Chem., 17, 2758 (1978). 



REACTIVITY OF POLYSTYRENE -BOUND BIPYRIDINE AND THE CATALYTIC 
POTENTIAL OF ITS TRANSITION METAL COMPLEXES 

Anton El A'mma September 21, 1978 

Since 1971* there has been a growing interest in catalysis 
by solid supported transition metal complexes. 1 The metal complex 
which is grafted onto a support can either be an established 
homogeneous catalyst or its metal center could acquire catalytic 
activity after the process of heterogenization. 

Crosslinked polystyrene is a commonly used solid support and 
numerous ligands have been anchored to this resin through covalent 
bonds. The resin which is functionalized with the analog of tri- 
phenylphosphine 2 " 4 is known to heterogenize potent homogeneous 
catalysts. However it has the following drawbacks: 1, air 
sensitivity of the phosphorus center; 2, lability of phosphine- 
metal linkage in some systems; 3, polymer crosslinking induced 
by the complex which modifies the primary structure of the support. 

I In order to bypass the above disadvantages, we have indepen- 
dently undertaken a study of the reactivity of a polymer bound 
bipyridine [(P)-Bipy] reported by Neckers et al. 5 and investigated 
the catalytic activity of some of its transition metal complexes 
towards hydrogenation. The choice of immobilized bipyridine stems 
from the fact that this ligand stabilizes a wide range of metal 
oxidation states and functions as a chelate. 

I ,The bonding properties of (p)-Bipy were studied by reacting 
it with a series of carbonyl containing metal complexes and then 
comparing the IR spectra (C-0 stretching region) of the. various 
polymer attached products with those of compounds derived when 
bipyridine undergoes the corresponding reactions. Polymer bound 
pyridine was used in similar reactions to ascertain the bidentate 
behavior of (p)-Bipy. Thus, (p)-Bipy was reacted with Mo(C0) 



that 




■Bipy functions as a bidentate ligand. 



Catalytic hydrogenations were observed with the following 
catalyst precursors-. (P)-BipyPtCl 2 , (P )-BipyPdCl 2 , polystyrene 
dispersed palladium metal and ( P)-BipyRh(PPh 3 )C1 . The system 
(P) -BipyPtCl 2 shows a high degree of selectivity towards the 
hydrogenation of terminal olefinic bonds. The material 
(P )-BipyPdCl 2 reduces a wide variety of coordinating and non- 
coordinating substrates (with iscmerization where applicable) 
and polymerizes methyl acrylate. The activity of this catalyst 
can be attributed in part to the presence of metallic palladium. 
Polystyrene dispersed palladium was found to be a factor of 
2-3 times slower (rate of H 2 uptake) than 5% Pd/C in hydrogenating 
p-benzoquinone and nitrobenzene. This system (? )-BipyRh(PPh 3 ) CI 



• 



is even more effective in hydrogenating internal and terminal 
double bonds than (p)--BipyPdCl 2 . Tentative mechanisms are dis- 
cussed for the activity of both (p)-BipyPtCl 2 and (p)-BipyRh(PPh 3 )ci. 
All of the above catalyst precursors have been reused without any 
loss of activity. 

In summary., the use of immobilized bipyridine has facilitated 
the understanding of the coordination sphere of the grafted metal 
complex and generated some novel hydrogenation catalysts. 



Referenc es : 

1. J. C. Bailar, Cat. Rev. Sci. Eng. , 10, 17 (197^). 

2. R. H. Grubbs and L. C. Kroll, J. Am. Chem. Soc., 93, 3062 (l97l) 

3. R. H. Grubbs, L. C. Kroll and E. M. Sv/eet, J. Mol. Sci. Chem., 
A7(5), 1047 (1973). 

k. J. P. Collman, L. S. Hegedus, M. P. Cooke, J. R. Norton, 
G. Dolcetti, and D. N. Marauarct, J. Am. Chem. Soc. , 94, 
1789 (1972). 

5. R. J, Card and D. C. Heckers, J. Am. Chem. Soc., 99, 7733 (1977) 



• 



57 
Synthesis and Stereodynamics of Tetrairidiumdodecacarbonyl 

Derivatives 
Gordon F. Stuntz (Final Seminar) May 16, 1978 

It has been proposed that transition metal cluster compounds may 
serve as models of a hetereogeneous surface. 1 The mobility of coordinated 
carbon monoxide and the C-H bond scission of organic compounds are 
two processes common to both metal clusters and metal surfaces which 
are of current interest. 

New synthetic techniques have been developed for the preparation 
of iridium carbonyl cluster compounds. Reductive carbonylation of the 
readily available Tr(I) species Ir (CO) 2 (p-toluidine)Cl under moderate 
conditions (5 atm CO, 90°C) provides Ir u (CO)i 2 in high yield. 2 
The addition of one-fourth equivalent of a phosphorus ligand to the 
reaction mixture allows the direct preparation of mono- and di-phosphorus 
ligand derivatives of Ir u (C0)i 2 . 3 Furthermore, it was found that 
the replacement of carbonyls in Ir*(CO)i 2 by a wide variety of ligands 
can be facilitated through the use of Me 3 N0*2H 2 as an oxidative 
decarbonylation reagent. In this manner complete series of Ir 4 (CO)i 2 
derivatives, having up to four carbonyls replaced, have been prepared 
for a number of isonitrile and phosphorus ligands. 

NMR ( 13 C, 3X P, 1 H) studies have shown that the phosphorus ligand 
derivatives adopt structures with three bridging carbonyl ligands. 1 * 
In contrast, the isonitrile derivatives generally adopt structures with 
only terminal carbonyls. Variable temperature 13 C NMR studies indicate 
that carbonyl site exchange in the mono-substituted derivatives occurs 
by rapid interconversions between carbonyl bridged and unbridged 
structures'* as originally proposed by Cotton for Rh u (CO)i 2 and 
Co 14 (CO) 12 . 5 

The direct reaction of Ir u (C0)i 2 with 1, 5-cyclooctadiene resulted 
in products arising from partial dehydrogenation of the organic 
ligands. Subsequent crystal structure determinations (Dr. C. G. Pierpont) 
revealed several unusual bonding modes for the organic moieties in two 
of these complexes. In Ir 7 (CO) x 2 (C 8 Hi 2 ) ( C e Hx i ) (C 8 H X ) the C B H U 
ligand is bound to an edge as a vinyl moiety by one a and one tt bond. 
The CsHio ligand bridges one triangular face and is bound with two 
a and one tt bond. 6 In contrast, the C e H 10 unit in Ir u ( CO) 5 (C 8 Hi 2 ) 2 (C 8 Hi. ) 
was found to be inserted into an Ir-Ir bond to give a Ir u C 2 pseudo- 
octahedral cluster framework. 7 Several simple 1, 5-cyclooctadiene 
derivatives of Ir u (C0)i 2 , (Ir u (CO) i 2 -2 X (C 8 H 1 2 ) x , x=l-3) were prepared 
under milder conditions and the intermediacy of these compounds in the 
formation of Irj,(CO) s (C 8 Hi 2 ) 2 (C e Hio) was established. The results Imply 
that the formation of Ir 4 (CO) 5 (CsHx 2 ) 2 (C 8 Hi ) involves two sequential 
oxidative additions of olefinic C-H bonds, followed by the insertion 
of the resultant acetylenic unit into an Ir-Ir bond giving the Ir u C 2 
pseudo-octahedral framework. 7 



58 



The mechanisms of carbonyl mobility and C-H bond scission 
determined for these iridium clusters provide models for related 
processes which occur on metal surfaces. 

References 

1. E. L. Muetterties, Science, 196, 839 (1977). 

2. G. F. Stuntz and J. R. Shapley, Inorg. Nucl. Chem. Lett., 12, ^9 
(1976). 

3. G. F. Stuntz and J. R. Shapley, Inorg. Chem., 15, 199^ (1976). 

1J . G. F. Stuntz and J. R. Shapley, J. Am. Chem. Soc . , 9_9, 607 (1977) 

5. F. A. Cotton, Inorg. Chem., 5, 1083 (1966). 

6. C. G. Pierpont, G. F. Stuntz, and J. R. Shapley, J. Am. Chem. Soc 
100 , 616 (1978). 

7. G. F. Stuntz, J. R. Shapley, and C. G. Pierpont, Inorg. Chem., in 
press . 



3 



INCLUSION BY MOLECULAR CAVITIES: APPLICATIONS 
Les Butler October 12, 1978 

The flexibility offered by the use of molecular cavities has 
excited much interest in spite of the added complications involved 
with their application. Zeolites, 1 silicates, 2 and graphite 3 
represent long-standing examples of exploitation of cavities for 
various applications. The quest for solution phase cavities is 
satisfied, at least in part, with cyclodextrins , "* ' 5 sugar deriva- 
tives, and cryptands , 6 ' synthetic multicyclic heteroatom compounds. 
The purpose of this seminar is to review some properties and sel- 
ected examples of the use of molecular cavities, some of which are 
large enough to contain substituted aromatics and others which can 
only include small molecules and metal ions. 

SMALL CAVITIES 

Most of the research with cryptands has dealt with the bi- 
cyclic compounds such as N (CH2CH2OCH2CH2OCH2CH2 ) 3N denoted as 
[2.2.2], although other topologically different derivatives are 
known. Early research showed that [2.2.2] prefers to include 
the potassium cation while [3.3.2] exhibited nearly equal for- . 
mation constants for potassium and rubidium cations. The observed 
stability was described in terms of the number and type of binding 
sites leading to the "cryptate effect" while the selectivity was 
seen to be a function of the fit of cavity size with the metal 
cation size. 8 The temperature dependence of the complex formation 
showed that the change in enthalpy reflected the selectivity of 
a particular cryptand. 3 In methanol, both selectivity and stability 
of the included complexes were enhanced. 8 ' 9 It was observed that, 
in methanol, the alkaline-earth complex with the greatest stability 
also possessed the greatest activation enthalpy for dissociation. 
Similar characteristics were also observed with alkali cations 11 
and it was suggested that the transition state for formation of the 
complex closely resembles the state of the reactants . l ° ' 1 1 The 
energy of desolvation of the metal cation would be compensated by 
interaction with the complex. 

Recent applications of small cavities have included the 
preparation of inert lanthanide cryptates, 12 stabilization of 
the sodium anion, 13 ' 11 * and use in forming isolated counter-ions. 15 

LARGE CAVITIES 

A convenient source of large cavities is provided by the 
cyclodextrins which are a-l,4-linked cyclic oligomers of D- 
glucopyranose (a, 3, and y refer to 6 , 7 and 8 glucose units, 
respectively) which form a hollow, hydrophobic, truncated cone 
ringed by twelve secondary hydroxyl functions at the large end, 
and six primary hydroxyl functions at the other end. 1 G A wide 
variety of molecules can be included into cyclodextrins resulting 
in many suggestions for the driving force for inclusion. Nuclear 
magnetic resonance 17 and crystal structure determinations 18 have 
been employed to examine the nature of the binding and orientation 
of included benzoic acid and sodium benzoate. These studies have 



shown that in both cases the acid function is located inside the 
cavity. A comprehensive model of the thermodynamics of inclusion 
of aquated apolar substrates was recently reported. l 9 

The applications of cyclodextrins have been numerous : the 
resolution of chiral sulfinyl compounds, 20 the preparation of 
Vitamins Ki and K 2 , 21 and catalytic hydrolysis of amides. 22 
Furthermore, the substrate binding by enzymes can be modeled 
with cyclodextrins or derivatives. 23 ,21t 

References 

1. P. A. Risbo d and D. M. Ruthven, J. Am. Chem. Soc, 100 , 
4919 (1978). 

2. J. M. Adams, J. A. Ballantine, S. H. Graham, R. A. Laub, J. H. 
Purnell, P. I. Reid, W. Y. M. Shaman and J. M. Thomas, Angew. 
Chem. Int. Ed. Engl., 17, 282 (1978). 

3. J. E. Fischer and T. E. Thompson, Physics Today, July 36, 1978. 

4. M. L. Bender and P. W. Griffiths, Adv. Cat., 23^ 209 (1973). 

5. M. L. Bender and M. Komiyama, "Cyclodextrin Chemistry," 
Springer-Verlag, Berlin, Heidelberg, New York, 1978. 

6. J. M. Lehn, Struct. Bonding (Berlin), 16 , 1 (1973). 

7. J. M. Lehn, Accts. Chem. Res., 11, 49 (1978). 

8. J. M. Lehn and J. P. Sauvage, J. Am. Chem. Soc, 9J7, 6700 (1975) 

9. E. Kauffmann, J. M. Lehn and J. P. Sauvage, Helv. Chim. Acta, 
59_, 1099 (1976) . 

10. V. M. Loyola, R. Pizer and R. G. Wilkins, J. Am. Chem. Soc, 99 , 
7185 (1977). 

11. B. G. Cox, H. Schneidner and J. Stroka, J. Am. Chem. Soc, 100 , 
4746 (1973) . 

12. 0. A. Gansow, A. R. Kauser, K. M. Triplet, M. J. Weaver, and 
E. L. Yee, J. Am. Chem. Soc, 9_9, 7087 (1977). 

13. F. J. Tehan, B. L. Barnett and J. L. Dve, J. Am. Chem. Soc, 
9_6, 7203 (1974) . 

14. J. L. Dye, M. R. Yemen, tl. G. DaGue, and J. M. Lehn, J. Chem. 
Phys. , S_S, 1665 (1978) . 

15. R. G. Teller, R. G. Finke, J. P. Collman, H. B. Chin and R. Bau, 
J. Am. Chem. Soc, 99_ 1104 (1977). 

16. P. C. Manor and W. Saenger, J. Am. Chem. Soc, 9_6, 3630 (1974). 

17. R. J. Bergeron, M. A. Channing and K. A. McGovern, J. Am. Chem. 
Soc, 100 , 2878 (1978). 

18. K. Harata, Bull. Chem. Soc, Jpn. , 50^, 1416 (1977). 

19. I . Tabushi , Y. Kiyosuke, T. Sugimoto and K. Yamamura, J. Am. 
Chem. Soc, 100 , 916 (1978). 

20. M. Mikolajczvk and J. Drabowicz, J. Am. Chem. Soc, 100 , 2510 

(1978). 



21. I. Tabushi, K. Fujita and H. Kawakubo, J. Am. Chem. Soc, 
99 , 6456 (1977) . 

22. M. Komiyama and M.. Bender, J. Am. Chem. Soc, 99, 8021 (1977) 

23. R. Breslow, J. B. Doherty, G. Guillot and C. Lipsey, J. Am. 
Chem. Soc, 100 , 3227 (1978). 

24. Y. Matsui, T. Yokoi and K. Mochida, Chem. Lett., 1037 (1976). 






Intervalence Electron Transfer in Mixed Valence Compounds 
Bruce C. Bunker Final Seminar October 19, 1978 

Experiments aimed at understanding tne energetics and rates 
of intervalence electron transfer in a variety of mixed-valence 
compounds have been undertaken to try to obtain a aata base for 
testing current theories regarding more general electron transfer 
phenomena. These theories can be applied in areas of chemistry and 
physics ranging from electrochemistry to semiconductor science. 
In particular, an attempt has heen made to check the validity and/or 
the limitations of the Hush theory , which relates the energy of 
the intervalence transfer band observed in the electronic spectrum 
to the energy of the thermal barrier to intervaj_ence electron trans- 
fer within the mixed- valence compound. 

Part of the research effort has been directed at studying 
mixed-valence compounds wnich show behavior which might be indi- 
cative of a breakdown in the Hush tneory. One such compound is the 
mixed-valence ^2>3l oxiaation state of the aimer ^-pyrazine-bis 
(pentaammineruthenium) tosylate, wnich contains one Ru(ll) and one 
Ru(III). Since Creutz and Taube first reported the synthesis of 
this aimer. in 1969, it has been the subject of a great deal of in- 
terest ana controversy in the literature. It exhibits an inter- 
valence transfer band which nas neither the band width nor the sol- 
vent dependence predicted on the Dasis of tne Hush theory. The 
many attempts which have oeen reported to experimentally determine 
the rate of intervalence electron transfer have j.ed to ambiguous 
results. 

To further our understanding of this compound and the Hush 
theory, we conducted experiments to help us formulate a description 
for the molecular orbitals involved in the electron transfer and to 
determine the rate of intervalence electron transfer between the two 
metal centers. Variable temperature magnetic susceptibility results 
on the [p,3jpyr aimer couplea with low temperature EPH results for 
ootn the [2,3] pyr ana D>»3Jpyr aimers indicate that the odd electron 
in the aimer resiaes in the (d ,a ) orbital set on the Ru(ll) 
center ana is not in a molecular oroital which is aelocalized over 
both metals as pos.aj.atea by Hush . Variable temperature stuaies 



indicate that in solution the rate of electron transfer from the 

Ru(II) center to the Ru(lII) center is fast on the jMMR time scale 

of 1<P sec from room temperature uown to -80 C. It is slow on the 

_q 
EPR time scale of 10 sec in the solid, state at all temperatures 

below -50 G. If outer coordination sphere effects are assumed to 
be negligible, then these &PR and jflMR results can be used to bracket 
the magnitude of the thermal energy barrier to between 3.4 and 6.7 
£cal/mole. This is in agreement with the Hush tneory prediction of 
4.5 kcal/mole, which is cased on the energy of the IT band. However, 
it may be that in the solid state, outer coordination sphere effects 
dominate, and that lattice effects can locic the odd electron onto 
one valence site. 

The other part of this research effort concerning mixed- 
valence compounds has been focused on trying to develop better ex- 
perimental techniques for studying the intervaience electron trans- 
fer phenomenon. The technique which has been investigated most ex- 

5 6 

tensively is time domain reflectometry , or TDR. ' . This technique 

can be used to study the dielectric relaxation properties of sam- 
ples in the frequency range from 10 Hz up to 10 Hz. The charge 
oscillations associated with intervaience electron transfer can 
give rise to such dielectric relaxation. Analysis of TDR data can 
yield the relaxation frequency associated with the relaxation pro- 
cess, which can be used to calculate the rate of electron transfer. 
Variable temperature TDR data can be used to calculate the magni- 
tude of the thermal energy barrier to electron transfer by using 
the Arrhenius equation. 

TDR data has ceen obtained for a sample of europium sulfide 
which analyzes as Eu.S,-. Analysis of the data indicates that the 
rate of electron transfer in this mixed valence compound is 2.2 x 10 
sec at room temperature, and that the magnitude of the thermal 
energy barrier to the electron transfer process is around 920 cal/mole 
This barrier corresponds almost exactly to the band gap energy 
associated with the semiconductivity of the compound. This has been 
determined from IR measurements, where the edge of the band gap ab- 
sorption feature can be observed at around 320 cm" . These results 
represent the first reported use of time domain reflectometry on 
solid i imples, and 'he first time the technique has been used to 






8 



determine the rate of electron transfer in a mixed valence compound. 

Other mixed-valence compounds have also been studied. Mag- 
netite, or Pe^O,, exhibits a rapia relaxation process which appears 

to be fast on the TDR time scale. This result is consistent with 

7 
Mossbauer results which have been obtained for the compound . 

Soluble Prussian blue exhibixs a relaxation process which is ob- 
servable in the TBR, but is slow on the TDK time scale of 10~ sec, 

Q 

which also agrees with the Mossbauer results . No relaxation is 
observed for solid U (NK^) t-Ru.) -W^ tosylate. This is consistent 

with the EPR study which indicates that the odd electron in the 
dimer is localized on one metal center in the solid state. The 
mixed-valence trimer Fe 2+ Fe 2 5+ 0(CH 3 COO) 6 (H 2 0) 5 , reported by Gol'- 
danskii to exhibit dynamic electron transfer on the Mossbauer time 
scale, has also been studied. It does not exhibit the expected di- 
electric relaxation in the TDK, nor do several other mixed-valence 
oligamers which have been studied. It is currently not known whether 
This means that the TDK technique is only sensitive to electron trans- 
fer in extended lattice systems, or that literature estimates of 
electron transfer rates are incorrect for tnese systems. 

References 

1. N. Hush, Prog. Inorg. Chem. , 8, 391 (1967) 

2. C. Creutz and H. Taube, J. Am. Chem. Soc, ^1, 3988 (1969) 

3. B. Bunker, R. S. Drago, 1). N. Hendrickson, R. M. Richman, and 
S. L. Kessell, J. Am. Chem. Soc, 100, 3805 (1978) 

4. J. Beattie, N. Hush, and P. Taylor, Inorg. Chem., 1£, 992 (1976) 

5. M. J. C. van Gemert, Phillips Res. Repts. , 28, 530 (1973) 

6. R. H. Cole, J. Phys. Chem., 22, 1459, 1469 (1975) 

7. R. Bauminger, et. al., Phys. Rev., 122, 1447 (1961) 

8. A. Ito, M. Suenaga, and K. Ono, J. Chem. Phys., 48, 3597 (1968) 

9. V. I. Cxol'danskii, et. al., Dokl. Akad. Nauk. S33R (Phys. Chem.), 
212, 1063 (1973) 



PHOTO-OXIDATION OF INORGANIC COMPOUNDS 
WITH APPLICATION TO SOLAR ENERGY 

Richard W. Wegman October 26, 1978 

Thermal conversion and quantum conversion are two fundamental 
ways for converting solar energy into useful energy. A thermal sys- 
tem utilizes a collector that transforms light energy into heat. l 
The heat is transferred to a thermal storage system for use at a 
later time. Quantum conversion utilizes the available solar energy 
to induce a photochemical or photoelectric process in the absorbing 
material. In a photochemical process the absorber is activated to 
an excited state by the absorption of light. The energy gained by 
the absorber is lost by deactivation of the excited state through 
radiative and non-radiative decay processes. 2 An energy storage 
system coupled to the excited state provides an alternative deacti- 
vation pathway by directing the excess energy into a process that 
produces a stable fuel. 3 The efficiency of the energy storage process 
is limited by certain thermodynamic and kinetic parameters of the 
absorbing system. The thermodynamic limitations include: 

1) The maximum chemical potential difference achievable between 
the ground and excited state; 1 * and 

2) Power drainage into the storage system and its effect on 
the photo-generated chemical potential. 5 

The kinetic limitations arise from the inherent microreversibility of 
the system. 6 

The photo-dissociation of water into 2 and H 2 is a very attractive 
system for solar energy conversion. 7 Hydrogen is an excellent fuel 
and the source is cheap and plentiful. Unfortunately, high energy 
photons, not abundant in solar radiation, are required for direct 
photo-decomposition of water. Chemical systems that photo-catalytically 
decompose water by utilizing lower energy photons are therefore necessary 
One of the oldest known systems involves the photo-oxidation of Ce(III) 
to Ce(IV) in dilute perchloric acid solution. The process begins with 
the photo-excitation of Ce(III) to *Ce(III). The *Ce(III) ion reduces 
H + resulting in the production of H 2 gas and Ce(IV). The system is 
cyclic because Ce(IV) is capable of oxidizing water to regenerate H + 
and Ce(III) . 

A similar system involves the photo-oxidation of Cu(I) to Cu(II) 
in the presence of H + . 9 This process is non-cyclic because Cu(II) 
cannot oxidize water to regenerate starting materials. A mechanism 
for H 2 production in this system has recently been proposed. * ° 

■ m I 

The photo-excitation of Ru(bpy) 2 . to *Ru (bpy) a and the conse- 
quent reduction of water by *Ru(bpy)f f has been suggested as a 
plausible method for water decomposition. 11 Apparently, for kinetic 
reasons, the electron transfer process is too slow to compete with 
excited state decay. Recently, Meyer 1 2 has reported a working 



10 



2 + 



photo-electrochemical cell utilizing Ru(bpy)g . The cathode is a 
solution containing Ru (bpy) g + and Co(C 2 0i h )3 3 ~ in 1 N H2SCU. The anode 
is a solution containing Fe(II) and Fe(III) 2 also in 1 N H2SOi». 
Photolysis_of the cathode produces *Ru(bpy)3 which is oxidized by 
Co(C20it)3 3 to give Ru(bpy)l + . A current is produced by the oxi- 
dation of Fe(II) by Rufbpy) 3 ,*. 

The photochemistry of di-nuclear metal complexes is of current 
interest and recently a rhodium 13 and several molybdenum di-nuclear 
compounds 1 "*' 15 have been shown to stoichiometrically reduce H + to H2 
during photolysis. 

The systems developed thus far have inherent problems which render 
them inadequate for the practical production of fuels. These problems 
include: 

1) High energy excitation wavelengths; 

2) Low quantum yields in the conversion process; 

3) Non-cyclic systems. 

Research oriented towards the engineering of systems capable of working 
at a practical level is necessary. 

References 

1. J. Richard Williams, Solar Energy , Technology and Applications , 
Ann Arbor Science Publishers, Inc., 1977. 

2. J. R. Bolton, J. Solid State Chem. , 22, 3 (1977). 

3. J. R. Bolton, to be published. 

4. R. T. Ross and Ta-Lee Hsiao, J. Appl. Phys., 4_8, 4783 (1977). 

5. R. T. Ross, J. Chem. Phys., 4_6, 4590 (1967). 

6. M. Almgren, Photochemistry and Photobiology , 2_7, 603 (1978). 

7. V. Balzani, L. Moggi , M. F. Manfrin, F. Bolletta and M. Gleria, 
Science, 189 , 852 (1975) . 

8. L. J. Heidt and A. F. McMillan, J. Am. Chem. Soc, ]j±, 2135 (1954). 

9. D. D. Davis, G. K. King, K. L. Stevenson, E. R. Birnbaum, and 
J. H. Hageman, J. Solid State Chem., _22, 63 (1977). 

10. G. Ferraudi, Inorg. Chem., 17, 1370 (1978). 

11. C. Creutz and N. Sutin, Proc. Nat. Acad. Sci., USA, 7_2, 2858 (1975). 

12. B. Durham and T. J. Meyer, J. Am. Chem. Soc, 100 , 6286 (1978). 

13. K. R. Mann, N. S. Lewis, V. M. Miskowski, D. K. Erwin, G. S. Hammond, 
and H. B. Gray, J. Am. Chem. Soc, 99, 5525 (1977). 



11 



14. D. K. Erwin, G. L. Geoffroy, H. B. Gray, G. S. Hammond, E. I. 
Solomon, W. C. Trogler and A. A. Zagars, ibid., 9_9, 3620 (1977). 

15. W. C. Trogler, D. K. Erwin, G. L. Geoffroy and H. B. Gray, ibid . , 
100, 1160 (1978) . 



28 



LIGAND EXCHANGE STUDIES OF METHYLCOBALAMIN AND 
MODEL METHYLATOCOBALT (III) CHELATE SYSTEMS 



Pamela Milton December 5, 1978 

Alkylatocobalt (III) chelate systems, containing planar or 
nearly planar ring systems, have been extensively investigated 
as models for alkylcorrinoid systems. Although a considerable 
amount of equilibrium data exists for base substitutions and ex- 
changes of alkylcorrinoid systems, the only kinetic study of 
methylcobalamin thus far reported involves a spectrophotometric 
investigation of the cyanation reaction of methylcobalamin in 
alkaline medium. 1 The value of 0.028 sec -1 derived as the rate 
constant at 25°C for displacement of dimethylbenzimidazole (Dmbz) 
by cyanide in methylcobalamin 1 is inconsistent with the observation 
of fast exchange in the 2 H and l 3 C NMR spectra of methylcobalamin 
at room temperature. 2 ' 3 

By protonation of methylcobalamin to the half-equivalence 
point and utilization of methanol as solvent, NMR spectra were 
obtained in the intermediate and stopped exchange region for Dmbz 
dissociation from methylcobalamin. Activation parameters for the 
dissociation of Dmbz from methylcobalamin were determined from 
detailed fitting of 1 3 C and ! H NMR line shapes in the CH3-C0 region. 
An Arrhenius plot based on the combined data, which were .in good 
agreement, yielded AHT = 11.1 ± 0.6 kcal/mol, AG? = 12.7 ± 0.1 
kcal/mol, and AST = -5.9 ± 2.4 cal/(K mol) . The rate of Dmbz dis- 
sociation in 50% methanol-water was found to be essentially the 
same as in neat methanol. The AHT for Dmbz dissociation is signifi- 
cantly lower than expected, based on comparison with AHT for 
dissociation of l-( 2- trif luoromethylphenyl) imidazole (CF3lm) from 
the methyl derivative of heptamethyl ester of cob (III) yrinic acid 
(methylcobester) . 4 This lower value may be ascribed to repulsive 
steric interactions which hinder the approach of Dmbz to cobalt. 

In an attempt to establish a firmer basis for comparing the 
interaction of CF 3 Im and Dmbz with differing cobalt centers, the 
kinetics of Dmbz dissociation from methylcobaloxime was examined. 
Activation parameters for CF3I111 dissociation from methylcobaloxime 
have been reported previously by Guschl and Brown. 1 * The axial 
coordination site of methylcobaloxime should be much less hindered 
sterically than that of methylcobalamin. Two sets of variable 
temperature NMR data were collected for solutions containing equi- 
molar Dmbz and benzonitrile adducts with methylcobaloxime in 
nitrobenzene. Unfortunately, rather large uncertainties are 
associated with the values of AHT and AST calculated for Dmbz 
dissociation from methylcobaloxime due to the rather limited 
temperature range available for study. Thus, no firm conclusions 
can be made for this system with regards to differences in acti- 
vation enthalpies for Dmbz and CF3I1T1 exchange with methylcobaloxime. 
However, relative values of AGT and rate constants, obtained in 
this study and a previous study 4 , indicate that the labilities of 
several heterocyclic base complexes with MeCbx decrease in the order 



29 



FPyr > Dmbz > CF 3 Im. 

An investigation of ligand exchange reactions of methylcobester 
was undertaken in an effort to obtain comparative kinetics infor- 
mation for the interaction of bases with methylcorrinoid complexes 
in nonpolar, non-interacting solvents. Establishment of NMR two- 
site exchange systems for methylcobester derivatives proved to be 
difficult. Only for the case of CF3I111 adduct was the slow exchange 
region clearly established. It was possible, nonetheless, to ex- 
tract valuable information from the base concentration dependence 
and temperature dependence of the NMR spectra of various derivatives, 
such as benzimidazole, 3-f luoropyridine, and CF3lm. Implications of 
these studies with regards to cobalamin binding to proteins and 
enzymes and its coenzyme activity will be discussed. 

References 

1. I. P. Rudakova, T. A. Pospelova, V. A. Borodulina-Shvets , 
B. I. Kurganov, and A. M. Yurkevich, J. Organometal. Chem. , 
6_1, 389 (1973) . 

2. J. D. Brodie and M. Poe, Biochemistry, 11, 2534 (1972). 

3. T. E. Needham, N. A. Matwiyoff, T. E. Walker, and H..P. C. 
Hogenkamp, J. Amer. Chem. Soc, 9_5, 5019 (1973). 

4. R. J. Guschl and T. L. Brown, Inorg. Chem., 12, 2815 (1973). 



30 

NITROGEN-14 NUCLEAR QUADRUPOLE RESONANCE SPECTRA OF 

COORDINATED PYRIDINE 

Gerald V. Rubenacker December 14, 1978 

The number and variety of ! 4 N NQR data available for compounds in 
which a nitrogen atom is coordinated directly to a metal ion or other 
Lewis acid is quite limited. Difficulties in the measurement of low 
frequency (3 to 0.2 MHz) l 4 N resonances for coordinated nitrogen are 
avoided by using an adiabatic demagnetization in the laboratory frame 
double resonance experiment. 1 By this method, the protons in a sample 
are first allowed to approach equilibrium alignment in a high magnetic 
field; next, the same is removed to zero field, where it is irradiated 
at a known frequency and subsequently returned to high field where a wide 
line NMR experiment determines the magnitude of the proton magnetization. 
Should the zero field irradiation frequency approach a ! "*N quadrupole 
resonance line, energy is resonantly absorbed by the nitrogen nuclei. 
Some of this energy is transferred to the proton system by level crossing, 
which in turn reduces the proton magnetization relative to that obtained 
when no energy is absorbed by the nitrogen nuclei. By repeating this 
cycle, a spectrum of irradiation frequency versus proton magnetization 
exhibits peaks corresponding to the N resonances. 

Nitrogen-14 NQR data for pyridine-Lewis acid complexes have been 
obtained and interpreted in terms of nitrogen orbital occupancies using 
a modified Townes-Dailey model. 2 For each quadrupole coupling constant 
and assymetry parameter obtained for pyridine nitrogens, a donor orbital 
population, Oq.o., can be calculated. It is assumed that the nitrogen is 
sp 2 hybridized when coordinated, and that the tt and N-C a orbital popula- 
tions are given by: 

a = a o + A(2 - a D0 ) 

b = b o + B(2 - o DQ ) 

respectively, where a Q and b Q are the respective free ligand populations, 
and A and B are constants calculated from the NQR data. That is, the 
populations of the it and N-C a orbitals are allowed to increase in pro- 
portion to the loss of electron density from the lone pair orbital. 

Pyridine complexes with lithium, zinc(II), cadmium(II), mercury(II), 
silver (I) and iron(0), pyridine-halogen compounds and pyridinium salts 
among other Lewis acid centers are discussed in terms of the effect of 
coordination on the nitrogen donor orbital of pyridine. Correlations 
of this donor orbital population with infrared absorption bands, nitrogen- 
acid bond lengths and halogen NQR are also described. 

References : 

1. D. E. Edmonds, Phys . Reports, C, 2_9, 233 (1977). 

2. Y.-N. Hsieh, G. V. Rubenacker, C. P. Cheng and T. L. Brown, 
J. Am. Chem. Soc, 99, 1384 (1977). 



31 

KINETIC ISOTOPE EFFECTS OF OXIDATIVE ADDITION OF HYDROGEN 

TO METAL COMPLEXES 

David J. Blumer January 23, 1979 

Introduction 

Since the discovery of the oxidative addition reaction of H 2 , O2 , 
CH3I, and many other species to transition metal complexes, 1 much effort 
has focused on determining whether this is an analogous reaction to the 
chemisorption of a gas on a metal surface. As more refinements of work 
from both sides of the problem with respect to both surface processes 
and homogeneous catalysis come to light, certain similarities seem to 
prevail . 

In 1965, Chock and Halpern 3 measured the kinetics of the oxidative 
addition of H 2 , 2 , and CH 3 I to trans-Ir (CO) CI (P (C 6 H 5 ) .3 ) 2 . They also 
reported a kinetic isotope effect of 1.22 at 30°C for the ratio of the 
rate of addition of H 2 to the addition of D 2 . The present work was aimed 
at a systematic study of the kinetic isotope effect of this reaction as 
a prototype of catalytic activation of H 2 , as well as development of more 
sophisticated experimental methods and theoretical analysis. 

Theory 

Theoretically, the kinetic isotope effect can be described within 
the transition state theory as the ratio of statistical mechanical 
partition function for each of the isotopic species. This is commonly 
broken down into the product of the mass-moment of inertia (MMI) , the 
excitation term (EXC) , and the zero-point energy (ZPE) terms. In organic 
chemistry, it is common (and usually accurate) to ignore the MMI and EXC 
terms since they are nearly 1.0. In inorganic reactions, this may be 
very dangerous, as will be pointed out. In general, simplistic arguments 
concerning the amount of bond-breaking and bond-making in the transition 
state for inorganic reactions should be viewed with scepticism, since 
this study has shown that even state-of-the-art theoretical calculations 
cannot make such concrete statements. 

Calculations 

Computer calculations were undertaken to explore the effects of 
having three vibrational modes contributing to the reaction coordinate, 
the effects of quantum-mechanical tunnelling, the temperature dependence, 
the transition state geometry, and the transition state force field on 
the kinetic isotope effect. 4 Many models of the transition state were 
investigated, ranging from a simple three-atom system, H H, to more 

V 

complex models such as L. '**»$, 



'"'NT 

•IV 



32 



Apparatus 

In order to measure the kinetic isotope effect, an apparatus was 
designed and constructed which can measure the rate of uptake or release 
of a gas in a reaction very accurately. The system is automated with 
the time and pressure data recorded on a cassette tape, which is read 
into a PDP-11 computer for kinetics analysis and display. The rate of 
reaction with each isotopic gas was measured and then the ratios of the 
rates taken to obtain the kinetic isotope effect. 

Results and Conclusions 

The kinetic isotope effect of the oxidative addition of hydrogen to 
Ir (CO) CI (P (CsHs ) 3 ) 2 in toluene was temperature dependent and ranged from 
1.05 to 1.33 at 26.77°C and 50.0°C, respectively. This was interpreted 
in terms of the transition state properties, which must include a 

hindered rotation of the H H moiety and two bending vibrations of the 

triangular M-H 2 group, in addition to the contributions mentioned above. 

References 

1. J. Halpern, Ace. Chem. Res., 3_, 386 (1970). 

2. G. A. Somorjai and D. W. Blakely, Nature, 258 , 580 (1975). 

3. P. B. Chock and J. Halpern, J. Am. Chem. Soc . , 8_8, 3511 (1966). 

4. W. E. Buddenbaum and P. E. Yankwich, J. Phys. Chem., 7^, 3136 

(1967); G. J. Wei and P. E. Yankwich, J. Chem. Phys., 60, 3619 (1974) 



33 



SUPER-IONIC CONDUCTIVITY IN NON-STOICHIOMETRIC OXIDES 



Stanley A. Roth February 15, 1979 



Introduction 

Interest in solids with ion conducting properties has increased 
due to their potential use as solid electrolytes and as electrodes 
in energy conversion systems. Ionic conductivity arises as the 
result of imperfections in a crystal structure. Transportation 
of ions occurs by interchange between normal sites and defect sites 
in the crystal lattice. Frenkel and Schottky defects are in- 
trinsic lattice defects and will be present in any pure ionic 
crystal above K. Statistical thermodynamics shows that in NaCl, 
at room temperature, there are on the average one defect for each 
10 6 ions [1-4] . 

A defect can also be caused by the presence of a foreign ion 
in a normal lattice site. If the impurity has a different charge, 
the production of a vacancy is required to keep electrical neutrality 
within the crystal. Non-stoichiometry can result if either an 
excess of cations or an excess of anions is present. Beta-alumina, 
an example with mobile cations, is a sodium conductor used in Na/S 
batteries. Calcia stabilized zirconia, an example with mobile 
anions, is an oxygen conductor for hydrocarbon-air fuel cells and 
for oxygen concentration electrodes. 

Beta-Alumina 

An idealized crystal structure was determined in 1937 by 
Beevers and Ross to have the formula Na 2 -HAlaO 3 . The crystal 
is composed of spinel blocks bound together by loosely packed 
layers containing Na + and 0~ 2 ions [5]. The crystal structure 
was refined in 19 71 and found to have approximately 29% excess 
sodium in the conduction band and a formula of Na]_ +x Al]_]__ x /3O17 , 
such that 0.15 < x < 0.30 [6]. 

It was found that sodium could be exchanged for other cations 
in the beta-alumina structure. Ion exchange is preformed at 300°- 
350 °C from the molten nitrate salt. Tracer diffusion experiments were 
carried out to determine the ionic conductivity of the various 
cations. The fact that Na + beta-alumina exhibits the highest 
conductivity probably indicates that it is the ideal size conductor [7] 

Techniques done to study the ionic conduction band in beta- 
aluminas include: x-ray diffraction [5,6], neutron diffraction [8], 
nuclear magnetic resonance [9], infrared and raman spectroscopy 
[10,11], and electron paramagnetic resonance [12]. The general 
consensus is that at elevated temperatures the conducting cation 
is not localized in a specific lattice site but moves through the 
lattice with a disorder pattern which resembles that of a two- 
dimensional liquid. 



34 



In recent years beta-alumina has been developed into an 
electrolyte for use in a sodium sulfur battery [13] . One cell 
designed by the Ford Motor Co. was operated for over two years 
and was fully recharged in excess of 9 800 times. 

Calcia Stabilized Zirconium Oxide 

At elevated temperature and when doped with 10% to 20% 
calcium oxide, zirconium oxide forms in the fluorite crystal 
structure and shows abnormally high oxygen ion conductivity. 
The Ca +2 ion replaces the Zr +U ion in the crystal and electrical 
neutrality requirements cause the creation of an oxygen anion 
vacancy [14], Because this structure has so many 'holes' in the 
crystal lattice, rapid ionic diffusion is expected. 

Various aspects of ionic conduction have been studied. 
These include effects of composition, temperature, as well as 
oxygen pressure on the oxygen ion conductivity. It was found 
for (CaO) q ^ -j_5 (Zr0 2 ) o 85 that tlie calcia concentration at maximum 
conductivity increased from about 13 mol% at 1000 °C to about 
15 moll at 1400°C [15] . 

Calcia stabilized zirconia electrolytes have been used in 
many different ways in recent years. It has found use in high 
temperature fuel cells due to its noncorrosive, as well as 
chemical, thermal and mechanical stability properties [16]. 
It has also found a place in electrodes used to measure the 
oxygen concentration in both high-temperature gases and liquid 
metals [17,18]. 

References 

1. R. A. Laudise, The Growth of Single Crystals ; Prentice-Hall, 
Inc., Englewood Cliffs, New Jersey (1970); Chapter 1, 
Section 3, "Imperfections in Crystals". 

2. L. Mandelcarn, Non-S toichiometric Compounds ; Academic Press, 
New York (1964); Chapter 1, Section 3, "Departures from the 
Classical Lattice" . 

3. W. S. Fyfe, Geochemistry of Solids ; McGraw-Hill Book Co., 
New York (1964); Chapter 12, "Defects in Crystals". 

4. S. Geller (Ed.), Solid Electrolytes ; Topics in Applied 
Physics Series (1977) ; Chapter 5, J. K. Kennedy, "The 
Beta-Aluminas"; Chapter 6, W. L. Worrel, "Oxide Solid 
Electrolytes" . 

5. C. A. Beevers and M. A. S. Ross, "The Crystal Structure 
of Beta Alumina", Z. Krist., 9_7, 59 (1937). 

6. C. R. Peters, et al., "Refinement of the Structure of 
Sodium Beta-Alumina", Acta Cryst., B27, 1826 (1971). 



35 



7. Y. F. Y. Yao and Kummer, "Ion Exchange and Rates of 
Diffusion in Beta-Alumina", J. Inor. Nucl. Chem., 29 , 
2453 (1967). 

8. W. L. Roth, "Crystallography of Superionic Conductors", 
Trans. Am. Cryst. Assn., JUL, 51 (1975). 

9. W. Bailey, et al., "NMR Study of Na + Motion and Distri- 
bution in Beta-Alumina", J. Chem. Phys., 6j4, 4126 (1975). 

10. S.J. Allen and J. P. Remeika, "Measurement of the 
Attempt Freq. for Ion Diffusion in Ag and Na Beta-Aluminas", 
Phys. Rev. Letters, 33, 1478 (1974) . 

11. L. L. Chase, C. M. Hao, G. D. Mahan, "Raman Scattering from 
Na and Ag in Beta-Alumina", Solid State Commun., JL8, 401 

(1976) . 

12. D. Gourier and J. Antoine et al . , "EPR Study of Cu + in the 
Conduction Plane of Beta Aluminas", Phys. Status Solid, 41(2) , 
423 (1977) . 

13. J. T. Kummer and N. Weber, "Sodium Sulfur Secondary Battery", 
Proc. Power Sources Conf., 2JL_, 37 (1967). 

14. R. C. Garvie, "Cubic Fields in the CaO-Zr0 2 System", J. Am. 
Ceram. Soc, 51, 553 (1968). 

15. T. H. Etsel and J. N. Flengas, "The Electrical Properties 
of Solid Oxide Electrolytes", Chem. Rev., 70^, 339 (1970). 

16. C. S. Tedmon, et al . , "Cathode Materials and Performance in 
High-Temperature Zirconia Electrolyte Fuel Cells", J. 
Electrochem. Soc, 116 , 1170 (1969). 

17. J. Weissbart and R. Ruka, "A Solid Electrolyte Fuel Cell", 
J. Electrochem. Soc, 109 , 723 (1962). 

18. D. Yuan and F. A. Kroeger, "Stabilized Zirconia as an Oxygen 
Pump", J. Electrochem. Soc, 116, 594 (1969). 



36 






POLYOXOANION CHEMISTRY 



Peter Dimas • February 20, 1979 



Occurrence and Structure 

A prominent feature of the aqueous chemistry of molybdate, 
tungstate and vanadate ions is their polymerization upon acidifi- 
cation to form large polyoxometallate anions [1] . The basic 
building unit of these anions is an octahedron of six oxygens 
surrounding each metal atom. Clusters of these octahedra are 
-linked by sharing apices, edges, or, less often, faces [2]. 
Polyoxoanions which contain only oxygen, the metallate atom, 
and possibly hydrogen, are termed isopoly ions. Heteropoly ions 
are those structures in whicn a heteroatom or various functional 
groups have been incorporated [3] . 

Current Research Motivation 

Much current interest in polyoxoanions is motivated by their 
diverse applications as heterogeneous catalysts [3]. Furthermore, 
polyoxoanions resemble fragments of close-packed oxide lattices 
and the behavior and properties of such anions may be used as a 
model for metal oxide catalyst-substrate interactions [4,5]. 

A clear understanding of the solution dynamics and transfor- 
mations of polyoxoanions is being sought in order to determine the 
mechanistic principles governing formation and reactivity of 
various polyoxoanions [6] , and to develop useful homogeneous 
catalysts. 

Also of interest is the ability of polyoxoanions to undergo 
facile, multi electron reduction to isostructural , Class II mixed- 
valence species which are called 'heteropoly blues' in recognition 
of their strong absorption in the red region of the visible 
spectrum [2] . 

Polymerization Mechanisms [6] 

Two fundamental constructional principles, the addition and 
condensation mechanisms, have been proposed. These mechanisms 
account satisfactorily for the particular species formed as well 
as for the rapid rates of formation observed. 

Dynamic Solution Behavior 

Investigation of two hexamolybdoorganoarsonates [7] , 
(PhAs) 2 Mo 6 2u u ~ (A), and (PhAs) 2 Mo 6 2 5 H 2 u ~ (B) , has shown them 
to differ by a constiutional^ water molecule. Rapid solution 



37 



interconversion of the edge-linked A form to B, which contains a 
shared face, proceeds by insertion of a bridging water into the 
A structure in wet or aqueous solutions. Similar fluxional 
behavior is observed for the corresponding tungstate , but must 
be induced by protonation [8]. These systems suggest the impor- 
tance of face-sharing in polyoxoanion hydrolyses. 

The dynamic stereochemistry of a-MOsC^e 4- has been investigated 
by 17 FT-NMR [9]. The two tetrahedral molybdate subunits are 
found to reorient rapidly within the structure, either inter- or 
intramolecularly. A simple relationship between kinetic lability 
and low bond order is established and its implications concerning 
.reactivity of other polyoxoanions are discussed. An exchange 
experiment between M 18 C4 2 ~ and a-Mo 8 2 6 4 ~ might indicate whether 
the reorientation process is inter- or intramolecular. 

The facile interconversion of two Mo 8 2 6 U ~ isomers has been 
discovered [10] . An intramolecular rearrangement process has 
been suggested but experimental verification is required. 

The decomposition equilibrium between the labile Mo 5 0i 7 H 3 ~ 
anion to the products a,-Mo 8 26 u ~ and Mo 2 7 2 has been elucidated 
by IR [11]. The weakly bound M0 U 2 ~ unit and proton account for 
the observed lability. A mechanism has been proposed. 

The rate of 18 exchange between H 2 and Vio0 28 6 ~ has been 
investigated [12] . A mechanism involving partial fragmentation of 
the anion has been developed to explain the observed exchange 
equivalency of all twenty-eight oxygen atoms. 

References 

1. D. L. Kepert in "Comprehensive Inorganic Chemistry", Vol. 4, 
J. C. Bailar, Jr., H. J. Emeleus, R. Nyholm, and A. F. 
Trotman-Dickinson, Ed., Pergamon Press, Oxford, 1973, pp. 607-672 
General introduction. 

2. Weakly, T. J. R. : Structure and Bonding 18 , 131 (1974). A 
review of structural and electronic aspects. 

3. G. A. Tsigdinos in Topics in Current Chemistry , Vol. 76, 

M. J. S. Dewar, et al . , Ed., Springer-Verlag , Berlin, 1978, 
pp. 1-65. A useful review of preparations, structures, 
properties, and uses. 

4. R. K. C. Ho and W. G. Klemperer, Polyoxoanion supported 
organometallics, J. Amer. Chem. Soc, 100 , 6772 (1978). 

5. V. W. Day, M. F. Frederich, R. S. Liu, and W. G. Klemperer, 
Polyoxymolybdate-hydrocarbon interactions, J. Amer. Chem. Soc, 
101, 491 (1979) . 



38 



6. K. H. Tytko and 0. Glemser, Adv. Inorg. Chem. Radiochem. , 
19 , 239 (1976). Describes various techniques used in 
investigating solution equilibria, evaluates measurements, 
discusses formation mechanisms. 

7. W. Kak, L. Rajkovic, M. T. Pope, C. Quicksall, K. Matsumoto, 
Y. Sasaki, Soluiton interconversion of heteropoly anions 
that differ by a constitutional water molecule, J. Amer . 
Chem. Soc, 9_9, 6463 (1977). 

8. S. H. Wasfi, W. Kwak, M. T. Pope, K. M. Barkigia, R. J. Butcher, 
C. 0. Quicksall, Protonation-induced dynamic stereochemistry, 

J. Am. Chem. Soc, 100 , 7786 (1978). 

9. V. W. Day, M. F. Frederich, W. G. Klemperer, W. Shun, 
Dynamic Stereochemistry of a-Mo 8 2 6 4 ~, J. Amer. Chem. Soc, 
99^, 952 (1977) . 

10. W. Shun, W. G. Klemperer, Interconversion of isomeric a- 
and 3-MO802 6 1 *" ions, J. Amer. Chem. Soc, 98_, 8291 (1976). 

11. M. Filowitz, W. G. Klemperer, W. Shum, Characterization of 
the pentamolybdate ion, Mo 5 0i 7 H 3 ~, J. Amer. Chem. Soc, 100 , 
2580 (1978). 

12. R. Kent Murmann and K. C. Giese, Mechanism of 18 exchange 
between water and Vi 28 6 "/ Inorg. Chem., ±1_, 1160 (1978). 

General Reference: 

W. G. Klemperer, V. 7 . Shum, K. C. Ho, 17 NMR of Polyoxometallates . 
I., Inorg. Chem., 18, 93 (1979). 



/ 



39 



ASYMMETRIC HYDROGENATION USING WILKINSON-LIKE CATALYSTS 
Sharon A. Brawner February 27 , 1979 



The synthesis of optically active compounds with high optical 
purity is a challenging problem. A common route in the prepara- 
tion of optically active compounds is the reaction of a chiral 
reactant with a prochiral substance [1] . Despite high optical 
yields, a major disadvantage of this method is the necessity for 
stoichiometric quantities of chiral reactants . Therefore it is 
desirable to prepare an asymmetric Catalyst which can produce 
chiral compounds of high optical purity without the use of chiral 
reactants. Heterogeneous techniques, which generally lacked 
stereospecif icity , resulted in low optical yields and irrepro- 
ducible results [2]. Stimulated by the development of Wilkinson's 
catalyst [3] and synthetic routes to optically active phosphines 
[4] , homogeneous catalysis has received much attention, in an 
attempt to optimize optical yields. One of the most effective 
asymmetric catalyst systems has been developed by generating the 
active species in situ from [Rh (diene) L n ] + X~ where L is an optically 
active phosphine. 

In the design of an efficient asymmetric catalyst the choice 
of ligand is crucial. Using a ligand chiral at phosphorus, 
Knowles was able to obtain modest optical yields in the hydrogena- 
tion of a-6-unsaturated carboxylic acids [5]. Attempts to improve 
optical yields were successful by modification of the phosphine 
with bulky substituents . Morrison [7] and Kagan [8] also obtained 
good asymmetric induction in the hydrogenation of a-acylaminoacrylic 
acids, using ligands in which the chirality resided on an aryl or 
alkyl substituent of the phospine and a chelating diphosphine, 
respectively . 

The nature of the substrate is also important in achieving 
high optical yields. Excellent results have been obtained in 
the hydrogenation of a-acylaminoacrylic acids. This has been 
attributed to their tridentate structure with coordination to 
rhodium occurring through the olefin, amide and carbonyl functional 
groups [9]. Another factor found to influence optical yield was 
the E or Z geometry of the olefin, the Z isomer being reduced with 
higher stereoselectivities and higher rates of hydrogenation [9, 10] 

R NHCOR 1 R COOH 

\ / \ / 

C=C r=C 

S \ / \ 

C 5 H 5 COOH C 6 H 5 NHCOR' 

(E) -isomer (Z) -isomer 

Geometric isomers of a-acylaminoacrylic acids 



40 



Although there has been a great deal of research involving 
asymmetric hydrogenation, little has been done to elucidate a 
detailed reaction mechanism. 31 P, ! H NMR and kinetic studies 
were conducted by Brown [11] and Halpern [12] on rhodium catalysts 
with chelating diphosphine ligands. The results were indicative 
of a mechanism in which the stereochemical course of the reaction 
is defined by olefin coordination to rhodium followed by oxidative- 



addition of hydrogen. 



3 1 



P NMR studies of rhodium catalysts with 



monophosphine ligands suggest that the primary reaction pathway 
in hydrogenation is dependent upon the structure of the phosphine 
[13] . 

The dependence of optical yield upon E or Z geometry of the 
olefin was examined by deuteration studies and *H NMR [14, 15]. 
E ■+■ Z isomerization was found to be responsible for this dependence 
Although addition-elimination is the major mechanistic pathway fo 
the isomerization of olefins, catalyzed by Rh(III) species [16], 
both Knowles [15] and Kagan [14] have rejected this pathway. 
For the catalyst [Rh(DI0f)]S + Kagan proposes pathways involving 
orthometalation of a phenyl ring on the substrate or the 
formation of an immonium intermediate [14] . For [RhL n S] where 
L is a variety of mono- and diphosphines , Knowles postulates a 
TT-allyl mechanism to account for the observed isomerization [15] . 



References 

1. H. C. Brown, N. R. Ayyangar, and G. Zweifel, J. Amer. Chem. 
Soc. , 8j>_, 397 (1964) . 

2. S. Akabori, S. Sakurai, et al., Nature (London), 178 , 323 (1956) 

3. J. A. Osborn, F. H. Jardine, J. F. Young, and G. Wilkinson, 
J. Chem. Soc, B, 1711 (1966). 

4. J. D. Morrison and W. F. Masler, Advan. Catal., 2_5, 81 
(1976) and references therein. 

5. W. S. Knowles and M. J. Sabachy, Chem. Coram. , 1445 (1968). 

6. W. S. Knowles and M. J. Sabachy, J. C. S. Chem. Coram., 10 (1972) 
J. D. Morrison, et al., J. Amer. Chem. Soc, 93, 1301 (1971). 



7. 
8. 

9. 

10. 
11. 



T. P. Dang and H. B. Kagan, J. Amer. Chem. Soc, 9_4, 6429 
(1972) . 

B. D. Vineyard, W. S. Knowles, et al. , J.- Amer. Chem. Soc, 
99_, 5946 (1977) . 

H. B. Kagan and G. Gelbard, Tetrahedron, 3_2y 233 (1976) . 

J. M. Brown and P. A. Chaloner, J. C. S. Chem. Coram., 321 (1978) 















41 



12. J. Halpern, D. P. Riley, et al., J. Amer. Chem. Soc, 99 , 
8055 (1977) . 

13. J. M. Brown, P. A. Chaloner and P. N. Nicholson, J. C. S. 
Chem. Comm., 646 (1978). 

14. C. Detellier, G. Gelbard, H. B. Kagan, J. Amer. Chem. Soc, 
100 , 7556 (1978) . 

15. K. E. Koenig and W. S. Knowles, J. Amer. Chem. Soc, 100 , 
7561 (1978) . 

16. R. E. Rinehart and J. S. Lashy, J. Amer. Chem. Soc, 86 , 
2516 (1964). 



42 



FERROFLUIDS 



Arrietta Walker March 27, 1979 

A ferrofluid is a colloidal suspension of magnetic particles 
in a liquid. The particles are stabilized by surfactant molecules 
adsorbed on the surface of the particles which are solvated by the 
carrier liquid [1]. The presence of a magnetic field does not 
separate the two phases, but causes the liquid to move with the 
particles as if the fluid were homogeneous. 

Ferrofluids were originally designed as a means of controlling 
fuel flow under the zero gravity conditions of free space. The 
fluids became commercially available in the early 1970' s. Applica- 
tions of the fluids cover a wide range of areas; use in loudspeakers, 
zero-leakage seals, damping liquids, magnetic ink, nonwearing 
electrical switches, artificial muscle [2], ore separation [3, 4], 
and energy converters [5] . 

The magnetic particles in the fluid are single domain ferro- 
magnets, which exhibit paramagnetic behavior. This phenomenon is 
referred to as superparamagnetism [6, 7, 8, 9]. The particles 
align their magnetic moments in the presence of an applied magnetic 
field, but are immediately randomized by thermal agitation when 
the field is removed leaving no residual magnetization. The 
particles differ from paramagnetic species in having a magnetic 
moment on the order of 10 4 u R per particle. 

Ferrofluids are synthesized by placing the magnetic compound, 
surfactant, and carrier liquid in a ball mill half filled with 
steel balls, and grinding for 1-6 months [1]. The properties of 
the ferrofluid can be altered by using different carrier liquids, 
such as water, f luorocarbons , silicate esters, kerosine, and 
paraffin wax. 

The ferrofluids have been characterized using a variety of 
physical methods including Mossbauer [10, 11, 12], magnetization 
measurements [13], and EPR [14]. 

References 

1. R. Kaiser and R. E. Rosensweig; Study of Ferromagnetic Liquid, 
NASA CR-1407, 1-97, 1969. 

2. R. Moskowitz; Ferrofluids: liquid magnetics, IEEE Spectrum 
12(3) , 53 (1975) . 

3. U. T. Andres, Magnetic Liquids, Materials Science and 
Engineering, 26, 269 (1976). 

4. S. E. Khalafalla and G. W. Reimers. Magnetic Levitation, 
AIME Transactions, 254, 193 (1973). 



43 



5. R. E. Rosensweig, Magnetic Fluids, International Science and 
Technology, p. 48, July, 1966. 

6. P. W. Selwood, Chemisorption and Bonding , Academic Press, 
New York, 1975. 

7. C. P. Bean and I. S. Jacobs, Magnetic Granulometry and Super- 
Paramagnetism. J. Appl. Phys., 27 (12) , p. 1448 (1956). 

8. C. P. Bean and J. D. Livingston, Superparamagnetism, J. Appl. 
Phys. Suppl. , 30(4) , p. 120S (1959). 

9. T. K. McNar, R. A. Fox, J. F. Boyle, Some Magnetic Properties 
of Magnetite (Fe 3 0i4) Microcrystals, J. Appl. Phys., 39(12), 

p. 5703 (1968) . 

10. T. Takada, M. Kiyama, and Y. Bando, Magnetic Properties of 
Several Iron Compounds Studied by the Mossbauer Effect, Bull. 
Inst. Chem. Res., Kyoto Univ., 47(4), p. 298 (1969). 

11. H. Winkler, H. J. Heinrich and E. Gerdau, Relaxation Phenomena 
in Ferrofluids, J. de Phys., C6-37, p. C6-261 (1976). 

12. S. Morup, H. Topsoe, and J. Lipka, Modified Theory for 
Mossbauer Spectra of Superparamagnetic Particles, J. de Phys., 
C6-37, p. C6-287 (1976) . 

13. R. Kaiser and G. Miskolczy, Magnetic Properties of Stable 
Dispersions of Subdomain Magnetite Particles, J. Appl. Phys., 
41(3) , p. 1064 (1970) . 

14. V. K. Sharma and F. Waldner, Superparamagnetic and ferrimag- 
netic resonance of ultrafine Fe30<+ particles in ferrofluids, 
J. Appl. Phys., 48(10), p. 4298 (1977). 



44 



METAL VAPORIZATION: A ROUTE TO SOME NOVEL ORGANO-TRANSITION 

METAL CHEMISTRY 

Debra S. Strickland April 5, 1979 

Since the first report [1] of the reaction of transition 
metal vapors with organic compounds, research into synthesis 
using metal vapors has steadily expanded and has been the source 
of much fascinating new chemistry [2-4]. This technique [5] in- 
volves the generation of gaseous free metal atoms under high vacuum, 
and the low temperature condensation of these atoms with a large ■ 
excess of an organic substrate vapor. Both static and rotating 
reaction vessels have been employed, and the latter has been 
adapted for the condensation of metal atoms into a cold solution 
of an involatile substrate in an inert solvent [6] . 

The vaporization method yields metal atoms in reactive high 
chemical potential states and, for this reason, has provided a 
convenient route to otherwise inaccessible organometallic systems. 
In particular, the incorporation of substituents possessing lone 
pairs of electrons into ir-complexed arene rings had not been 
possible via the conventional Fischer-Hafner synthesis. Using 
metal vaporization, McGlinchey and coworkers have succeeded in 
making many fluorinated bis (arene) chromium complexes which undergo 
nucleophilic substitution and proton abstraction reactions [7]. 
Nucleophilic organometallic carbanions have been observed to form 
fluorinated polychromaren.es [3] and have been used in the synthe- 
sis of bis (arene) chromium complexes containing a wide variety of 
organic substituents [9]. A new type of metal atom synthesis has 
been devised by Timms and coworkers [10] , in which potassium atoms 
are condensed into a solution of a metal halide and an arene, 
forming bis (arene) metal complexes. 

The oxidative addition of alky 1 -halogen bonds to Pd atoms 
yields products which are too thermally unstable to observe [11] . 
However, Klabunde and Roberts obtained a novel r\ 3 -benzylpalladium- 
chloride dirner [12] upon the cocondeiisation of benzyl chloride 
and Pd atoms. Further studies [13] showed that the presence of 
aluminosilicate crucible insulation material produced organic 
free radicals which caused destruction of this dimer, as well as 
the isomerization and polymerization of alkenes. 

Other recent examples of the applications of metal vapors in 
chemical 1 synthesis are numerous and include the synthesis of a 
new C-LsCo 2 metallocarborane cluster [14], the synthesis of Mi + (C0)t + - 
(hexafluoro-2-butyne) 3 clusters (M = Ni , Pd) [15], the reaction of 
1 , 3-butadiene with lanthanide atoms to form low-valent complexes 
[16], and the synthesis of (C^ F5 ) 2 Ni (n G -toluene) [17] which has an 
exceedingly labile arene ligand. Many compounds formed by the 
metal vapor route have proven to be useful catalysts [2] and 
transition metal -atoms themselves have been found to behave as 
efficient catalvsts in a number of isomerization, hydrogenation , 
disproportionation, and polymerization reactions [2,18]. 



45 



An important goal of organometallic chemistry in recent 
years has been the activation of saturated hydrocarbons. Skell 
and coworkers have reported the oxidative addition of C-C- and 
C-H bonds of necpentane and isobutane to Zr atoms at low tem- 
peratures [19] , an apparently unusual behavior among the metal 
atoms studied up to the present. Their hypothesis is supported 
by the nature of the products produced upon hydrolysis with 
D 2 0. • 

Klabunde and coworkers have reported that codeposition of Ni 
vapor with weakly complexing solvents allows the formation of very 
reactive high surface area metal slurries and metal powders [2 0] . 
There is evidence [21,22] that alkane cleavage occurs at low 
temperature to form solvent fragment-stabilized Ni clusters. The 
metal particles can be tailored to have high catalytic activity 
or selectivity in hydrogenation and isomerization reactions [22] 
■and highly dispersed catalysts have been obtained when metal atom 
solutions, formed by this method, are allowed to permeate catalyst 
supports [23] . 

The last decade has been an exploratory phase for preparative 
scale metal atom chemistry and there is growing interest in using 
the technique on a larger scale [24]. The variety of organo- 
transition metal compounds which have been prepared by the metal 
vaporization route illustrates the wide scope that this technique 
offers. Because of the accessibility of new chemistry, as well 
as its unique experimental advantages, metal vaporization will 
undoubtedly continue to gain importance as a widely-used synthetic 
approach. • _• 

Referen ces ■ • . 

1. P. L. Timms, J. Chem. Soc. Chem. Commun., 1033 (1969). 

2. P. I. . Timms and T. W. Turney, Adv. Organomet. " Chem. , 15 , 53 

■ (1977) ..... \ 

3. M. J. McGlinchey and P. S. Skell, in Cr yochemi stry , M. • 
Moskovits and G. A. Ozin, eds . , Wilev-Inter science, New York, 
1976. Chap. 5. 

4. K. J. Klabunde, Ace. Chem. Res., 8, -393 (1975). 

5. P. L. Timms, in Cryochemi str y , M. Moskovits and G. A. Ozin, 
eds., Wiley-Inter science, New York, 1976, Chap. 3. 

6. R. Makenzie and P. L. Timms, J. Chem. Soc. Chem. Commun., 
6 5 (19 7 4). 

7. M. J. McGlinchey and T. S. Tan, j. Am. Chem. Sec, 9_8, 2271 (1976) 

8. A. /^garwal, M. J. McGlinchey, and T. S. Tan, J. Organomet. Chem., 
141, 8 5 (1977) . 



46 



N.. Hao and M. J. McGlinchey, J. Organomet. Chem., 165 , 225 (1979) 

P. N. Hawker, E. P. Kiindig, and P. L. Tiihms , J. Chem. Soc 
Chem. Comun., 73 (1978). 

K. J. Klabunde and J. S. Roberts, J. Organomet. Chem., 137 , 
113 (1977). 

J. S. Roberts and K. J. Klabunde, J. Am. Chem. Soc, 99 , 
2509 (1977) . 

K. J. Klabunde, T. Groshens, H. F. Efner, and M. Kramer, 
J. Organomet. Chem., 157, 91 (1978). 

G. J. Zimmerman, R. Wilczynski, and L. G. Sneddon, J. Organomet. 
Chem. , 15£, C29 (1978) . 

K. J. Klabunde, T. Groshens, M. Brezinski, and W. Kennelly, 
J. Am. Chem. Soc, 100 , 4437 (1978). 

W. J. Evans, S. C. Engerer, and A. C. Neville, J. Am. Chem. 
Soc. , 1_0_0, 331 (1978) . 

K. J. Klabunde, B. B. Anderson, M. Bader , and L. J. Radonovich, 
J. Am. Chem. Soc,. 100 , 1313(1978).. 

V. M. Akhmedov, M. T. Anthony, M. L. H. Green, and D. Young, 
J. Chem. Soc, Dalton Trans,', 1412 (1975). 

R, J. Remick, T. A. Asunta, and P. S. Skell, J. Am. Chem. Soc, 
101 , 1320 (1979) . 

K. J. Klabunde, II. F. Efner, T. 0. Murdoch, and R. Roppie, 
J. Am. Chem. Soc, 9_8, 1021.(1976).- 

S. C. Davis and K. J. Klabunde, J. Am. Chem. Soc, 100 , 5973 
(1978). ■ 

K. J. Klabunde, S. C. Davis, H. Hattori, and Y. Tanaika, 
J. Catal., 54^, 254 (1978). 

K. J. Klabunde, D. Ralston, R. Zoellner, H. Hattori, and 
Y. Tanaka, J. Catal., 5_5_, 213 (1978). 

W. Reichelt, Angew. Chem. Ind. Ed. Engl., 14, '218 (1975). 



47 



HOMOGENEOUS CATALYSIS OF THE WATER GAS SHIFT REACTION 



Robert Olsen April 12, 1979 



The water gas shift reaction (Eq. 1) is used extensively in 
industry today [1] to control the H2 :CO ratio in various reactions 
[2] . 

CO+H 2 O^H 2 +C0 2 (1) 

Presently, two main heterogeneous systems are used to catalyze the 
water gas shift reaction. Homogeneous systems are being studied 
to find more economical and thermodynamically favored processes. 

Metal carbonyls can be viewed as complexes containing carbon 
monoxide in an activated form. That is, a more positive charge 
on the carbonyl carbon makes it more susceptible to nucleophilic 
attack [3], Thus, metal carbonyls are logical candidates for use 
as homogeneous catalysts for the water gas shift reaction. 

Metal carbonyls react with water, presumably through a 
hydroxycarbonyl intermediate, to form CO 2 and a metal hydride 
[4-7] . The similarities between this reaction and the water gas 
shift reaction suggest a mechanism which can be used as a model 
system (Eq. 2) for the water gas shift reaction. 



II _ -co 2 

M y (CO) x + H 2 -^[HOCM y (CO) x _ 1 ] > [HM y (CO) x _ ± ] 

H 2° + C0 ■ 



^ OH + H-M (CO) , > M (CO) (2) 

z y x J. u y x 
-H 2 

This model can be used as a basis of comparison for the different 
proposed mechanisms. It appears that the mono- and polynuclear 
metal carbonyl systems [8-11], such as Fe(CO) 5 [11] and Ru 3 (CO)i 2 
[8-10], follow a reaction route similar to the proposed model. IR 
and NMR spectra of their reaction solutions indicate that a metal 
hydride is present. There is also evidence that a polyhydride 
species may be forming and that H 2 is reductively eliminated from 
this species in a rate-determining step. 

Other systems may well follow rather different pathways. In 
mononuclear metal carbonyl systems under high CO pressure [12] , 
supportive evidence has been found for CO displacement of the 
hydride as H~ from the metal center. The hydride could then react 
with water to form H 2 . The Group VI metal carbonyl systems are 



48 



found to lose activity after 20-25 turnovers [13] . This has been 
attributed to their forming a hydride-bridged dimer. In the Ptlo 
system [14], metal hydride formation is thought to occur via trans 
addition of water across the platinum center. In both the 
K 2 PtCli f /SnCli t /SnCl 2 [15] and [Rh(CO) 2 Cl] 2 [16] systems, hydride 
formation is thought to occur by way of H + addition to the metal 
center. Metal reduction would then take place in the C0 2 elim- 
ination step without any metal hydride formation. 

In conclusion, numerous transition metal complexes have been 
found to catalyze the water gas shift reaction. The proposed 
mechanisms differ, in minor respects from that shown in Equation 2. 
In all of these mechanisms, a hydroxycarbonyl intermediate is 
proposed from which C0 2 is expelled. 

References 

1. "Catalyst Handbook", Springer-Verlag, London, 1970. 

2. H. H. Storch, N. Golumbic, and R. B. Anderson, "The Fischer- 
Tropsch and Related Synthesis", Wiley, New York, NY, 1951. 

3. D. J. Darensbourg and M. Y. Darensbourg, Inorg. Chem. , 9_, 
1691 (1970) . 

4. E. L. Muetterties, Inorg. Chem., 4_, 1841 (1965). 

5. D. J. Darensbourg and J. A. Froelich, J. Am. Chem. Soc . , 99 , 
5940 (1977) . 

6. D. J. Darensbourg and J. A. Froelich, J. Am. Chem. Soc, 100 , 
338 (1978) . 

7. W. Hieber and H. Vetter, Z. Anorg. Allg. Chem., 217 , 145 (1933) 

8. R. M. Laine, R. G. Rinker , P. C. Ford, J. Am. Chem. Soc, 99 , 
252 (1977) . 

9. P. C. Ford, R. G. Rinker, C. Ungermann, R. M. Laine, V. Landis, 
S. A. Moya, J. Am. Chem. Soc, 100 , 4595 (1978). 

10. R. M. Laine, J. Am. Chem. Soc, 100 , 6451 (1978). 

11. H. C. Kang, C. H. Mauldin, T. Cole, W. Slegeir, K. Cann, 
and R. Pettit, J.. Am. Chem. Soc, 99^ 8323 (1977). 

12. R. B. King, C. C. Frazier, R. M. Hanes, and A. D. King, Jr. 
J. Am. Chem. Soc, 100, 2925 (1978). 



49 



13. D. J. Darensbourg and M. J. Incorvia, Inorg. Chem. , 18^ 18 
(1979) . 

14. T. Yoshida, Y. Ueda, and S. Otsuka, J. Am. Chem. Soc . , 100 , 
3941 (1978). 

15. C. H. Cheng and R. Eisenberg, J. Am. Chem. Soc, 100 , 5968 
(1978) . 

16. C. H. Cheng, D. E. Hendricksen, and R. Eisenberg, J. Am. Chem, 
Soc. , 99, 2791 (1977) . 



50 



STRUCTURAL STUDIES OF PROTEINS IN SOLUTION USING NMR 

AND PARAMAGNETIC PROBES 



James R. Stahlbush April 19, 1979 



The acquisition of structural information about proteins 
in solution is essential to our understanding of the biological 
activity of these proteins. In the past few years several re- 
searchers have turned to the use of paramagnetic probes in con- 
junction with nuclear magnetic resonance to obtain structural 
information about proteins in solution [1] . 

There are two types of paramagentic probes. Relaxation 
probes such as Gd(III), Mn(II) and organic spin labels broaden 
the NMR resonances of nuclei near the probe. The broadening, 
Av , is inversely proportional to r 6 , where r is the distance 
between the perturbed nuclei and the probe. Shift probes such 
as all the Ln(III) ions except Gd(III), La(III), and Lu(III) shift 
NMR resonances of nuclei near the probe. The shifting, A6 , is 
inversely proportional to r 3 and is also dependent on the angu- 
lar orientation of the perturbed nuclei in relation to the probe's 
magnetic axis [2]. Thus the degree to which a resonance is changed 
indicates the relative position of the perturbed nuclei with 
respect to the probe. 

The addition of a paramagnetic probe to a protein changes 
only a small percentage of the NMR resonances. Comparison of 
the perturbed spectrum to the normal spectrum (Paramagnetic 
Difference Spectroscopy [3] ) results in the assignment of NMR 
resonances to specific types of amino acid residues which are 
near the probe [4] . 

R. J. P. Williams and coworkers have extensively studied 
lysozyme using lanthanide ions as paramagnetic probes [4-6]. 
By studying the shift of methyl resonances perturbed by the shift 
probe Yb(III) at various pH values the lanthanide ions were 
proposed to bind simultaneously to the glutamic acid-35 and 
aspartic acid-52 residues [6]. The same binding site was ob- 
served in the crystal structure done by Imoto and coworkers [7]. 
This binding site is the active site of lysozyme which hydrolyzes 
glycosidic linkages of certain bacterial cell walls. Thus R. 
J. P. Williams and coworkers were able to map out the position 
of twelve amino acid residues within 15-20A of the active site 
[4-6] using the broadening probe, Gd(III), and various other 
lanthanides as shift probes. These NMR results were in good 
agreement with the crystal structure results [7] indicating that 
there are few differences between the two forms of lysozyme. 

Brewer and coworkers have studied the binding of l 3 C-enriched 
saccharides to Concanavalin A (ConA) using l 3 C NMR [8,9]. ConA 
has two metal sites per monomer, one Ca 2 site and one transition 
metal atom site. Comparision between the Ti values of the saccharide 
carbon atoms when bound to the paramagnetic Mn (II) -ConA derivative 



51 

versus the diamagnetic Zn(II)-ConA derivative indicated that 
the saccharide was bound approximately 10A from the transition 
metal ion site. This result contrasted with a preliminary crys- 
tal structure [10] estimate of 20-23A; however, a more recent 
crystal structure [11] agrees with the NMR results. 

Lanir and Navon have also used 13 C NMR to study the binding 
of acetate ions to carbonic anhydrase [12] which is a zinc con- 
taining enzyme. By substituting Mn(II) for the Zn(II), the dis- 
tance between the metal site and bound acetate ion was calculated 
to be 4 . 5&. The nine histidine resonances in the X H NMR spectrum 
were classified into three categories by comparing the resonances 
of the zinc analog to the paramagnetic cobalt (II) analog at var- 
ious pH values [13] . Three histidine residues were identified as 
ligands of the metal, three within the active site and three on 
the surface of the enzyme. 

Finally Dwek and coworkers [14,15] examined the active site 
of the Fv portion of the myeloma protein MOPC-315 using organic 
spin labels. Three conclusions were drawn from their studies. 
First, there are no major conformational changes upon the spin 
label binding to the antibody. Second, there are a relatively 
large number of aromatic amino acid residues at the binding site. 
Third, two of three histidine residues in the protein are near 
the binding site. The NMR results were compared to model building 
studies by Padlan and coworkers [16] . 

In conclusion, the use of paramagnetic probes with NMR 
allows structural information about proteins to be determined 
in solution. The major drawback of this technique is that pre- 
sently only amino acid residues with aromatic or methyl groups 
can be successfully indentified and assigned. 

References 

1. Morris, A. T. and Dwek, R. A., Some Recent Applications of the 
Use of Paramagnetic Centres to Probe Biological Systems Using 
Nuclear Magnetic Resonance, Quart. Rev. Biophys., 1_0, 421 (1977) 

2. Bleany, B. , Nuclear Magnetic Shifts in Solution Due to 
Lanthanide Ions, J. Mag. Resonance, £, 91 (1972) . 

3. Campbell, I. D., et al . , Resolution Enhancement of Protein 
PMR Spectra Using the Difference Between a Broadened and a 
Normal Spectrum, J. Mag. Resonance, 11, 172 (1973) . 

4. Campbell, I. D. , et al., Assignment of the *H NMR Spectrum of 
Proteins, Proc. R. Soc. Lond . A, 345 , 23 (1975). 

5. Campbell, I. D. , et al., Nuclear Magnetic Resonance Studies 
on the Structure of Lysozyme in Solution, Proc. R. Soc. Lond. 
A, 345, 41 (1975) . 



52 



6. Dobson, C. M. and Williams, R. J. P., Nuclear Magnetic 
Resonance Studies of the Interaction of Lanthanide Cations 
with Lysozyme, Jerusalem Symp. Quantum Chem. Biochem. , 9_, 
255 (1977) . 

7. Imoto, T, , et al., The Enzymes (ed. , P. D. Boyer) , New York: 
Academic Press, 1972, p. 666-868. 

8. Brewer, C. F. , et al., Interactions of Concanavalin A. 
Mechanism of Binding of a- and (3-Methyl D-Glucopyranoside 
to Concanavalin A as Determined by 1 3 C Nuclear Magnetic 
Resonance, Biochemistry, 1_2, 4448 (1973). 

9. Brewer, C. F., et al., 13 C NMR Studies of the Interaction 
of Concanavalin A with Saccharides, Adv. Exper. Med. Biol., 
5_5, 55 (1975) . 

10. Edelman, G. M. , et al., The Covalent and 3-Dimensional 
Structure of Concanavalin A, Proc . Natl. Acad. Sci., USA, 
69, 2580 (1972) . 

11. Hardman, K. D. and Ainsworth, C. F. , Structure of the 
Concanavalin A-a-Methyl-D-Mannopyranoside Complex at 6 A 
Resolution, Biochemistry, 15 , 1120 (1976) . 

12. Lanir, A. and Navon, G., NMR Studies of Two Binding Sites 

of Acetate Ions to Manganese (II) Carbonic Anhydrase , Biochim. 
Biophys. Acta, 341 , 75 (1974). 

13. Campbell, I. D. , et al., A Study of the Histidine Residues 
of Human Carbonic Anhydrase B Using 27 MHz Proton Magnetic 
Resonance, J. Mol. Biol., 90_, 469 (1974). 

14. Dwek, R. A., et al., Antibody-Hapten Interactions in Solution, 
Phil. Trans. R. Soc. Lond. B, 272 , 53 (1975) . 

15. Dwek, R. A., Structural Studies in Solution on the Combining 
Site of the Myeloma Protein MOPC 315, Contemp. Topics Mol. 
Immunol. , 6_, 1 (1977) . 

16. Padlan, E. A., et al., Model-Building Studies of Antigen- 
Binding Sites: The Hapten-Binding Site of MOPC-315, Cold 
Harbor Symp. Quant. Biol., 41, Part 2, 627 (1977). 



53 



SEMICONDUCTING RARE EARTH METALS 



Michael K. Kroeger April 25, 1979 



Semiconducting rare earth (RE) compounds have been known 
for less than twenty years. Early studies were plagued by a 
constant problem: purity. Extremely pure samples are required 
for meaningful measurements. Researchers are still unable to 
prepare stoichiometric pnictides but success has been obtained 
in the preparation of RE hexaborides and chalcogenides . 

Until recently it was thought that ferromagnetism was 
restricted to crystalline metals and alloys. However, many of 
the europium semiconductors have exhibited ferromagnetic 
ordering and other RE semiconductors have exhibited anti- 
ferromagnetic ordering. The interesting magnetic properties 
have made the RE chalcogenides and hexaborides heavily studied 
systems . 

Theoretical work has shown that hexaborides of RE 3 ions 
will be metallic while hexaborides of RE +2 ions will be semi- 
conducting [1] . Optical, magnetic, and electrical studies have 
shown that only EuB 6 and YbB 6 are semiconducting at room 
temperature [1,2]. EuB 6 is interesting because it is the only 
ferromagnetic hexaboride [1,3]. All other hexaborides are 
antiferromagnetic (LaB 6 is paramagnetic and YbB 6 and SmB 6 are 
nonmagnetic) . The only other semiconducting hexaboride is 
SmB6 which is a nonmagnetic semiconductor at low temperatures 
[4,5]. Mossbauer results show an isomer shift for SmB 6 which 
is between those for SmF2 and SmFs [4], Since SmB 6 is a semi- 
conductor, there can be no Sm +3 ion. It was thus postulated 
that SmB6 has two electronic configurations in its semiconducting 
phase: 4f 6 5d° and 4f 5 5d 1 [4]. 

The europium monochalcogenides are the most heavily studied 
of the RE compounds [6-11] . Only europium and thulium mono- 
chalcogenides exhibit any magnetic order. The europium 
monochalcogenides order both ferro- and antif erromagnetically 
while thulium telluride orders antif erromagnetically [7] . 
Samarium monochalcogenides are also very interesting because 
they undergo pressure induced semiconductor- to-metal phase 
transitions [5,12,13], The electronic structure of SmS is 
also similar to that of SmB 6 . 

RE sesquioxides and sesquisulf ides are also semiconductors 
with the interesting fact that they are also mixed-valence 
semiconductors [14,15]. Finally, Eu 3 Sit also exhibits mixed- 
valence semiconducting properties and appears to undergo 
conduction by an electron-hopping process [16,17]. 



54 



Although no longer seriously considered for practical 
semiconductor devices, RE compounds still provide very good 
semiconductor models due to their interesting and sometimes 
unique electric, magnetic, and optical properties. 



References 



1. J. P. Mercurio, et al . , J. Less-Common Met., 4J7, 175 (1976). 

2. M. Lalanne, et al., J. Less-Common Met., 41_, 181 (1976). 

3. Z. Fisk, Phys. Letters, 34A , 261 (1971) . 

4. R. L. Cohen, et al., J. Appl. Phys., 41_, 898 (1970). 

5. M. B. Maple and D. Wohllenben, A. I. P. Conf. Proc . , 18 , 
447 (1974). 

6. P. Wachter, Crit. Rev. in Solid State Sci., 3, 189 (1972). 

7. G. Giintherodt, Festkorperprobleme, 16_, 95 (1976). 

8. R. K. Ray, et al., Phys. Letters, 37A , 129 (1971). 

9. K. Kaskai, P. Kuivalainen, and T. Stubb, J. Appl. Phys., 
4_9, 1595 (1978) . 

10. R. Merlin, et al . , Solid State Commun., 2_2, 609 (1977). 

11. D. Hulin and C. a'la Guillaume, Solid State Commun., 25 , 
235 (1978). 

12. G. V. Lashkarev and L. A. Ivanchenko, J. Non-Cryst. Solids, 
8-10 , 670 (1972). 

13. F. Holtzberg, A. I. P. Conf., Proc, 18_, 478 (1974). 

14. G. V. Subba Rao, et al., J. Solid State Chem., 2, 377 (1970) 

15. P. Peshev, W. Piekarczyk, and S. Gazda, Mat. Res. Bull., 6_, 
479 (1971). 

16. H. H. Davis, et al . , J. Less-Common Met., 22^, 193 (1970). 

17. G. C. Allen, et al., J. Inorg. Nucl. Chem., 35, 2311 (1973). 



55 



SUPPORTED METAL CARBONYL CLUSTERS 



Alan Zombeck May 1, 1979 



As our energy demands increase and resources decrease, the 
development of more efficient chemical processes is essential. 
These new developments may reside in the discovery of new 
catalysts. Recently, there have been attempts to correlate 
relationships between homogeneous and heterogeneous catalysts. 
Studies of metal clusters have begun with the hope they may 
serve as models for metal surfaces [1,2]. Known homogeneous 
catalysts have been "heterogenized" by attaching these catalysts 
to polymeric supports [3,4]. These two concepts have been com- 
bined through the investigation of polymer-bound clusters. 
Supported metal carbonyl clusters offer several advantages. 
Interactions that lead to aggregation of the metal are minimized. 
Thus, after decarbonylation, it might be possible to obtain a 
coordinatively unsaturated species whose particle size may be 
controlled by using the appropriate cluster precursor. 

The reactivity of a metal carbonyl cluster with polymeric 
supports was studied using Rh 6 (CO) i 6 . Bassett and Smith in- 
vestigated the reaction of this cluster with silica gel [5] . 
They found the terminal carbonyl groups were oxidized by surface 
water or hydroxide groups. The original cluster could be re- 
generated with CO pressure. These workers also investigated to 
what extent the cluster could be decarbonylated while maintaining 
its integrity. K. L. Watters and coworkers studied the reactivity 
of Rh 6 (CO) i 6 on alumina [6], A gradual decarbonylation of the 
cluster is catalyzed by the alumina surface. The cluster most 
likely retains its integrity since it can be regenerated with CO. 
Reaction of Rh 6 (CO)i 6 with triphenylphosphine has been investi- 
gated in solution as well as with the polymer bound analog [7,8] . 
There is general agreement that the species present on phos- 
phinated polystyrene is Rh 6 (CO) i 3 (PPh 3 ) 3 where the cluster 
integrity has been maintained [8,9]. Jarrell and Gates investi- 
gated the IR and kinetics of this species for olefin hydrogenation . 
Cyclohexene hydrogenation was second order in olefin and half 
order in hydrogen. The ligands present on the functioning 
catalyst were inferred to be hydride, hydrocarbon, and polymer- 
attached phosphine groups. CO acted as an inhibitor in the 
catalytic reaction. The same workers also investigated the 
aggregation of the metal cluster using trace amounts of oxygen 
in the feed stream [10] . The phosphine groups were oxidized 
allowing the Rh to agglomerate and slowly forming a supported 
catalyst containing Rh crystallites about 20 A in diameter. 
These were characterized by IR and electron microscopy. Rh G (CO) i 6 
has also been immobilized onto phosphinated silica gel [11] . The 
tetranuclear Ir carbonyl cluster was found to maintain its integrity 
when bound to phosphinated polystyrene [12,13], 



56 



Interesting catalysts have been prepared from metal carbonyl 
clusters. Ichikawa studied the catalytic properties of 

[Cp 2 Ni 2 (CO) 2 ] and [Cp 3 Ni 3 (CO) 2 ] on silica gel [14]. ESR results 
suggest the trinickel complex maintained a three atom center on 
the silica after decarbonylation . The catalyst was active for 
the hydro formylation of ethylene in low yields, whereas a 
catalyst prepared from Ni(Cp) 2 was not. Ichikawa also prepared 
highly dispersed Pt aggregates by the pyrolysis of Pti 5 -Pt 3 anion 
clusters [15] . Pt aggregates were active catalysts for the 
dehydrocyclization of n-hexane . Systematic trends in selectivity 
of products were observed with the size of the cluster precursor. 
Basset and Ugo found hydrocarbon formation occurred with subse- 
quent heating of a metal carbonyl cluster with an inorganic 
support [16] . Surface water was confirmed as the source of H2 
through D 2 exchange. Soon thereafter, Ichikawa reported the 
synthesis of methanol and ethanol over catalysts prepared by 
pyrolysis of metal clusters on various metal oxides [17,18,19]. 
Catalytic properties of the supported metal crystallites were 
found to depend remarkably on the cluster precursor and the type 
of support. Increased catalytic activity was demonstrated by 
these catalysts when compared to conventional Rh metal catalysts. 
H. B. Gray and coworkers have prepared active hydro formylation 
catalysts by the photolysis of either Rlu (CO) 1 2 or Coi+(C0)i2 
solutions in the presence of polyvinylpyridine [20] . Rates were 
found to be comparable to those achieved by industrial Rh 
catalysts under more drastic conditions. Ichikawa as well has 
reported a hydroformylation catalyst using his pyrolysis method 

[21] . 

In summary, Rh G (CO) 1 6 was reacted with various polymeric 
supports to determine what reactions may occur between metal 
clusters and these supports. Novel catalysts have been prepared 
through polymer attachment of metal clusters. Activity of these 
catalysts was found to depend on the cluster precursor and the 
support. 

References 

1. E. L. Muetterties, Science, 196, 839 (1977). 

2. A. K. Smith and J. M. Basset, J. MoLCatal., 2, 229 (1977). 

3. Y. Chauvin, D. Commereuc, and F. Dawans, Prog. Polym. Sci., 
5, 95 (1977) . 

4. F. R. Hartley, P. N. Vezry, Advan. Organometal. Chem. , 15 , 
73 (1978) . 

5. J. M. Basset, et al., J. Organometal. Chem., 153 , 73 (1978). 

6. K. L. Watters, et al., Inorg. Chem., 14_, 1419 (1975). 

7. K. L. Watters, et al., Inorg. Chim. Acta, 15, 191 (1975). 



57 

8. J. P. Collman, et al., J. Am. Chem. Soc, 9_4, 1789. 

9. M. S. Jarrell and B. C. Gates, J. Catal . , 54_, 81 (1978). 

10. M. S. Jarrell, B. C. Gates, and E. D. Nicholson, J. Am. 
Chem. Soc, 100 , 5727 (1978). 

11. H. Knozinger and E. Rumpf, Inorg . Chim. Acta, 3_0, 51 (1978). 

12. B. C. Gates, et al . , J. C. S. Chem. Comm. , 540 (1978). 

13. J. R. Anderson, R. F. Howe, Nature, 268 , 129 (1977). 

14. M. Ichikawa, J. C. S. Chem. Comm., 26 (1976). 

15. Ibid . , p. 11. 

16. A. K. Smith, et al . , J. Am. Chem. Soc, 100 , 2590 (1978). 

17. M. Ichikawa, Bull. Chem. Soc, Jpn. , 51, 2268 (1978). 

18. Ibid . , p. 2273. 

19. M. Ichikawa, J. C. S. Chem. Comm., 566 (1978). 

20. H. B. Gray, et al . , in "Organometallic Polymers", ed. 

C. E. Carraher, J. E. Sheats, C. V. Pittman (1978) , pp. 155-164 

21. M. Ichikawa, J. Catal., 56, 127 (1979). 



58 

ELECTRON DYNAMICS OF METAL-QUINONE COMPLEXES 
Stephen L. Kessel May 3, 1979 



Quinones play an integral role in many biological electron 
transfer processes, particularly respiration and photosynthesis 
[1] . During these processes, quinones are reversibly reduced to 
semiquinone radical anions, with the concomitant oxidation of 
divalent metal centers. Para-quinones, as exemplified by ubi- 
quinone, have been found to transfer electrons via the semiquinone 
form from the high potential iron sulfur protein Hipip to the 
cytochromes [2]. Ortho-quinones , as totally reduced catechols, 
function as reducing agents in enzymatically catalyzed insertion and 
non-insertion reactions whereby O2 is reduced to H 2 [1,3]. 

A priori to the understanding of these electron transfer 
reactions is the fundamental apperception of the reduction- 
oxidation characteristics of the quinone and the nature of metal- 
quinone interactions. The redox function of the quinone has been 
well established [4], However, relatively few investigations of 
quinones reacting with divalent metal ions have been reported 
[4,5]. To elucidate the electronic and molecular structure of 
metal-quinone complexes, a series of para-quinones and ortho- 
quinones was reacted with M 11 (salen) , where M = Fe , Mn, Co and 
salen is N,N ' -ethylenebis (salicylideneimine) , and Fe^(TPP), 
where TPP is tetraphenylporphine . 

The reduction of para benzenoid quinones, such as p-benzo- 
quinone and duroquinone , to the dianion of the hydroquinone was 
accomplished with Fe 11 (salen) . This reaction, first reported by 
Floriani, yields bridged compounds of the type (salen) Fe^-^-Q- 
Fe III (salen) [6,7]. The quinone undergoes a two electron reduction 
to the dianion, as each ferrous ion is concurrently oxidized to 
the high spin trivalent state. These complexes are characterized 
as dimeric, square pyramidal systems exhibiting weak antiferro- 
magnetic exchange interactions with J values ranging from -0.2 cm -1 
to -5.9 cm -1 . Infrared data substantiate the reduction of the 
quinone while Mossbauer spectra confirm the oxidation of the iron. 
Several unhindered p-quinones have been reacted with Fe 11 (TPP) to 
form similar bridged systems displaying slightly larger J values 
[8], Mn 11 and Co^ 1 (salen) also react in this manner, but only with 
high potential quinones. 

Metal complexes ligated by semiquinones have been prepared 
by oxidative addition in a manner similar to the synthesis of 
para-quinone complexes. M 11 (salen) , where M = Fe , Mn, and Co, 
and various orthoguinones react via a one electron transfer 
process to form M^ 11 (salen) Q systems [9,10]. Complexes of 9,10- 
phenanthrenequinone, 3 , 5- di-tert-butyl-o-benzoquinone , and 1,2- 
naphthoquinone have been synthesized. Chemical and EPR data suggest 
that coordination of the semiquinone occurs through both oxygen 



59 



atoms [10,11], Mossbauer spectra clearly establish the oxidation 
state of iron as +3 high spin. Complete coupling of the radical 
anion electron, residing in a semiquinone it molecular orbital, 
with an iron d orbital electron results in a net S = 2 spin system 
exhibiting a room temperature U e ff of 4.85 B.M. A similar anti- 
ferromagnetic exchange is found for Mnm/salen/Q complexes. In 
the case of cobalt complexes, X-band EPR data confirm the oxidation 
of the cobalt, with the unpaired electron spin residing on the 
semiquinone moiety. Attempts to isolate a semiquinone coordinated 
to Fe 11 (TPP) proved unsuccessful. Reactions of various brtho- 
quinones with the ferrous porphyrin yielded a mixture of high 
spin ferric material and [Fe(TPP)] 2 0. While redox occurs, 
stabilization of the semiquinone is not accomplished, presumably 
because the stereochemical requirement for quinone coordination 
is not met. The semiquinone complexes disproportionate, forming 
2:1 metal-hydroquinone systems and free quinone. Magnetic 
susceptibility data on [Fe (TPP) ] 2 o-chloranil and [Fe(salen) ] 2 - 
o-chloranil suggest the two trivalent metal ions are coordinated 
to separate oxygen atoms . 

By reacting orthoquiones with Fe (CO) 5 and Cr (CO) 6 , a series 
of neutral tris-semiquinone complexes has been prepared [12] . 
As with the 1:1 metal-semiquinone systems, magnetic susceptibility 
data indicate a strong antiferromagnetic exchange interaction 
between metal d electrons and semiquinone unpaired electrons. 
Iron tris ligated with 3 , 5-di-tert-butyl-o-benzoquinone exhibits 
a magnetic moment of 2.90 B.M. at room temperature, suggesting 
complete coupling of the 3 ligand based electrons with 3 electrons 
derived from iron orbitals. Structural data on a related complex 
confirm the presence of semiquinone ligands and suggest a plausible 
pathway for extended intermolecular interactions. 

References 

1. "Inorganic Biochemistry", G. Eichorn, Ed., Elsevier Publishing 
Company, Amsterdam, 1973; R. H. Thompson, "Naturally Occurring 
Quinones", 2nd Ed., Academic Press, New York, 1971; M. 
Schnitzer and S. U. Khan, "Humic Substances in the Environment", 
Dekker, New York, 1972. 

2. F. J. Ruzicka, H. Beinert, K. L. Schepler, W. R. Dunham, 

and R. H. Sands, Proc. Natl. Acad. Sci. U.S., 72_, 2886 (1975); 
A. A. Konstantinov and E. K. Ruuge, Bioorg. Khim. , 3_, 787 (1977) 

3. "Free Radicals in 3iology", Vol. I & II, W. H. Pryor, Ed., 
Academic Press, New York, 1976; "Oxidases and Related Redox 
Systems", Vol. I & II, T. E. King, H. S. Masen and M. Morrison, 
Eds., John Wiley & Sons, New York, 1965. 

4. "The Chemistry of the Ouinonoid Compound", Parts 1 and 2, 
S. Patai, Ed., John Wiley and Sons, New York, 1974. 



60 



5. A. Y. Girgis, Y. S. Sohn, and A. L. Balch, Inorg. Chem. , 14, 
2327 (1975); A. A. Vlcek and J. Hazlik, Inorg. Chem., 6_, 
2053 (1967); G. N. Schrauzer, J. Araer. Chem. Soc, 8_2, 6420 
(1960) . 

6. C. Floriani, G. Fachinetti , and F. Calderazzo, J. Chem. Soc, 
Dalton, 765 (1973) . 

7. S. L. Kessel and D. N. Hendrickson, Inorg. Chem., 17_, 2630 (1978) 

8. S. L. Kessel and D. N. Hendrickson, manuscript in preparation. 

9. C. Floriani, R. Henzi, and F. Calderazzo, J. Chem. Soc, 
Dalton, 2640 (1972) . 

10. S. L. Kessel, R. Emberson, P. Debrunner, and D. N. Hendrickson, 
manuscript in preparation. 

11. V. Zelewsky and O. Haas, Proc. Int. Conf. Coord. Chem., 16th, 
2.256 (1974) . 

12. R. M. Buchanan, S. L. Kessel, H. H. Downs, W. B. Shorthill, 
C. G. Pierpont and D. N. Hendrickson, J. Amer, Chem. Soc, 
100 , 4318 (1978); R. M. Buchanan, S. L. Kessel, H. H. Downs, 
C. G. Pierpont, and D. N. Hendrickson, J. Amer. Chem. Soc, 
100, 7894 (1978) . 



61 

A MAGNETIC AND MOSSBAUER STUDY OF SOME MIXED VALENCE COMPOUNDS 
Karen Hassett Final Seminar May 7, 1979 



Magnetite, Fe30t+, is the mixed valence mineral commonly 
used as the material for compass needles. It was discovered 
by Briggs and Kjargaard [1] that when iron salts were reacted 
in a lignosulfonate matrix (lignin is a major component of 
wood) , an iron oxide was formed that had interesting magnetic 
properties. Magnetic and Mossbauer studies of several different 
samples of this material confirm that the microparticles formed 
are magnetite. These particles are small enough to consist 
only of single domains, and behave superparamagnetically. When 
observed on the very fast (^10~ 7 sec"^ ) Mossbauer timescale, the 
thermal magnetic relaxation of the magnetite is such that some 
of the particles show a superparamagnetic doublet, while other 
particles which relax more slowly give a complete magnetic hyper- 
fine spectrum. The ratio of these two types of hyperfine changes 
with the particle size and the temperature of the experiment. 
The average particle size is larger in those samples with a 
greater percent of iron. This is reflected by an increased 
magnetization and larger amount of magnetic hyperfine in the 
Mossbauer spectrum. 

The organometallic compounds diferrocenylketone (DFK) and 
triferrocenyldiketone (TFK) form mono- and dioxidized compounds, 
respectively. Unlike many ferrocene compounds studied, the 
relative intensities of the quadrupole split doublets (corres- 
ponding to the ferrocene and ferricenium portions of the 
compound) in these Mo'ssbauer spectra do not have integral 
ratios. This means that the two types of iron centers do not 
have identical absorption of gamma rays, i.e., their recoil 
free fraction (rff) is different. This is unusual since 
intermolecular vibrations generally make the most important 
contribution to the rff. Similar phenomena have been observed 
previously, but it was assumed in those cases that there was a 
migration of the electron density, causing the ratios of the 
peaks to change with temperature. It had been assumed that the 
recoil free fraction of ferrocene and ferricenium moieties would 
be similar in different compounds. Studies of some physical 
mixtures show that this is not the case. Since the rff of ferro- 
cene as a function of temperature has been determined [2] , the 
relative area of the ferrocene peaks in these mixtures was used 
to find the rff of three ferricenium salts. These were found to 
differ substantially from the values calculated for ferrocene. 

References 

1. William Scott Briggs and Niels J. Kjargaard, U. S. Patent 
4,019,995, Apr. 26, 1977. 

2. C. R. Hill, Ph.D. Thesis, University of Illinois, Urbana, 
111. (1978). 



62 



THE NATURE OF ACTIVE SITES IN CATALYTIC ZEOLITES 
Steven L. Suib May 11, 1979 

Zeolites are crystalline aluminosilicates consisting of an 
ordered framework structure forming a network of cavities inter- 
connected by pores of molecular dimensions. Positively charged 
ions and water molecules reside in these cavities and can move 
about freely during cation exchange and dehydration. 

There are a number of uses of zeolites in areas as diverse 
as pollution control, energy conservation and agriculture. Appli- 
cations in these areas range from oil spill cleanups to animal 
nutrition. Undoubtedly, however, the use of zeolites in petroleum 
refining is the most important and significant application of 
zeolites at this time. 

Despite the widespread use of zeolites a number of important 
questions regarding zeolitic behavior remain unanswered. The two 
most significant areas of doubt concern 

(1) The nature of the metallic sites in zeolites and 

(2) The role of protons in zeolites. 

An incomplete list of zeolite books [1-5] and review articles [6-10] 
is given in the list of references. 

The zeolite ZSM-5 is particularly interesting because it 
catalyzes the transformation of methanol into gasoline hydro- 
carbons [11-13]. This zeolite is not commercially available but 
can be synthesized [14]. Very little is known about ZSM-5 and 
therefore we have chosen to study its thermal, ion exchange, 
electronic and chemical properties. 

Europium exchanged zeolites are of interest because the rare- 
earth-exchanged zeolites are important as catalysts and because 
the crystallinity of the zeolites provides cation sites of a 
definite symmetry. A single crystal X-ray structure of europium 
exchanged zeolite A has been reported and the presence of Eu 1 *" 1 " 
has been postulated [15,16]. In addition, a process has been 
reported for thermolytically dissociating water using europium 
exchanged zeolites [17]. It is believed that the conversion of 
europium to the 2+ and 3+ oxidation states in important in this 
thermolytic cycle. By using Mossbauer [18], EPR, luminescence 
lifetime [19] and EXAFS [20] spectroscopy we investigated the 
nature of the europium sites in zeolites A, X, Y and ZSM-5 
as a function of hydration and chemical treatment. 



63 



A recent X-ray photoelectron spectroscopy study of zeolites 
has indicated that there are major differences between the surface 
and the bulk composition of zeolites [21] . Such a compositional 
differentiation may be important in determining the selectivity 
of a particular zeolite. Single crystal and powdered zeolites 
have been subjected to Auger electron spectroscopic analysis in 
order to investigate this problem. In addition, AES has been 
used to observe changes in metal ion concentrations on zeolites 
as a function of dehydration in hopes of observing metal reduction 
and migration. 

Finally, the isomerization of 1-butene catalyzed by various 
europium zeolites will be described. It is extremely important 
to study catalytic processes on well characterized systems. Sys- 
tematic observations of the catalysis of this reaction may lead 
to a greater understanding of the active sites in zeolites. Factors 
of importance for this reaction are silica-alumina ratio of the 
zeolite, extent of europium exchange, temperature of activation, 
activation pretreatment, structural integrity of the zeolite, 
cation migration, impurity levels and the reactivity of unexchanged 
zeolites. 



References 

1. J. A. Rabo, Zeolite Chemistry and Catalysis, ACS Monograph 171, 
American Chemical Society, Washington, D.C., 1976. 

2. J. R. Katzer, Molecular Sieves-II, ACS Symposium Series 40, 
American Chemical Society, Washington, D.C., 1977. 

3. F. A. Mumpton, Mineralogy and Geology of Natural Zeolites, 
Mineralogical Society of America, Short Course Notes, Volume 4, 
November 1977. 

4. D. W. Breck, Zeolite Molecular Sieves: Structure, Chemistry 
and Use, Wiley and Sons, New York, 1974. 

5. P. A. Jacobs, Carbogenionic Activity of Zeolites, Elsevier 
Scientific Publishing Company, Amsterdam, 1977. 

6. D. W. Breck, J. Chem. Ed., 4JL, 678 (1964). 

7. R. M. Barrer, Chem. Ind. , 1203 (1968). 

8. H. W. Haynes, Jr., Cat. Rev.-Sci. Eng., 17, 273 (1978). 

9. J. H. Lunsford, Cat. Rev.-Sci. Eng., 12, 137 (1975). 

10. P. B. Weisz, Chem. Tech., 498 (1973). 

11. N. Y. Chen and W. E. Garwood, J. Cat., 52, 453 (1978). 



64 



12. G. A. Mills and B. M. Harney, Chem. Tech., 26 (1974). 

13. P. D. Caesar, J. A. Brennan, W. E. Garwood and J. Ciric, 
J. Cat. , 5_6 274 (1979) . 

14. U. S. Patent, 3, 702, 886. 

15. R. L. Firor and K. Seff, J. Am. Chem. Soc, 100 , 976 (1978). 

16. R. L. Firor and K. Seff, J. Am. Chem. Soc, 100 , 978 (1978). 

17. P. H. Kasai and R. J. Bishop, Jr., U. S. Patent, 3, 963, 830. 

18. S. L. Suib, R. P. Zerger, G. D. Stucky, R. M. Emberson and 
L. E. Iton, The Oxidation States of Europium in Zeolites, 
submitted to J. Am. Chem. Soc. 

19. W. De W. Horrocks, Jr., G. F. Schmidt, D. R. Sudnick, 
C. Kittrell, and R. A. Bernheim, J. Am. Chem. Soc, 99 , 
2378 (1977). 

20. (a) A. H. Reis, Jr., L. E. Iton, G. K. Shenoy, T. I. Morrison, 
S. L. Suib and G. D. Stucky; SSRL Users Meeting, October 1978. 

(b) A. H. Reis, Jr., L. E. Iton, T. I. Morrison, S. L. Suib 
and G. D. Stucky, Sixth North American Meeting of the Catalysis 
Society, March 1979. 

21. J. Fr. Tempere , D. Delatosse, and J. P. Contour, Chem. Phys. 
Lett., 33, 95 (1975). 



65 



NUCLEAR QUADRUPOLE RESONANCE STUDIES OF 
METAL COMPLEXES OF BIOLOGICAL INTEREST 



Carol I. H. Ashby May 15 r 1979 



The importance of metal ions in biological systems has been 
well established [1] . Zinc plays an important role as a Lewis 
acid in metalloenzymes such as carboxypeptidase and carbonic 
anhydrase. It is, therefore, important to understand the nature 
of the interaction between zinc and biological model ligands such 
as imidazole, amino acids, and peptides. Nuclear quadrupole 
resonance spectroscopy [2] has proven to be a useful technique 
for the study of the interaction between diamagnetic metal ions 
and nitrogen donor ligands. 

The probe employed in this study is the quadrupolar nucleus , 
ll *N f for which 1=1. The separation between the quadrupolar energy 
levels is determined by the orientation dependent interaction 
of the nitrogen nucleus with its electronic environment. The 
electric field gradient (efg) , which determines the energies, may 
be factored into contributions arising from electrons in molecular 
orbitals and charges on neighboring nuclei. An approximate method 
formulated by Townes and Dailey [3] which ascribes the efg solely 
to nitrogen valence p-electrons has been successfully employed. 
Information about the bonding of nitrogen with metal ions may be 
obtained by relating efg data to populations of hybrid orbitals 
centered on nitrogen. 

Difficulties in detecting nitrogen quadrupolar transitions 
in the region from to 2 MHz may be overcome by use of the double 
resonance by level crossing (drlc) technique [4] . In this tech- 
nique, an abundant spin species such as protons is initially 
polarized in a large static magnetic field. During transit of 
the sample to a zero field region, the quadrupolar spin system 
equilibrates with the proton spin system by spin exchange during 
level crossing. The sample is irradiated with a radio frequency 
magnetic field in the zero field region and returned to the high 
field region, again undergoing level crossing exchange. The 
magnitude of the proton magnetization is then detected by a wide- 
line NMR experiment. If the irradiation frequency corresponds 
to a quadrupolar transition frequency, energy is resonantly ab- 
sorbed by the nitrogen nuclei and transferred to the proton 
spin system during the second level crossing. Transitions are, 
therefore, manifested as a decrease in the magnitude of the 
proton magnetization. 

The interactions of Zn and Cd with imidazole as a function 
of coordination environment have been studied using the drlc technique 
[5] . The nitrogens are assumed to be sp 2 hybridized, as were 
those of a previous pyridine study [6] . If it is assumed that 
the two N-C o-bonds are equivalent, the efg components may be 
expressed in terms of orbital populations, a, a, and b, which refer 



66 



to the populations at nitrogen of the donor or lone pair orbital, 
the TT-orbital, and the C-N a-bond orbital, respectively. Co- 
ordination to a metal should decrease the donor orbital population 
from that of the free ligand and concomitantly produce an inductive 
response in the other orbitals as follows : 

a = a + A(2 - a) 

b = b + B(2 - a) 
o 

when 2, a, and b are the populations in the reference compound, 
N-benzylimidazole. The values of the scaling factors, A and B, 
are based upon experimental NQR data. From the calculated values 
of the efg components as a function of a, it is possible to relate 
experimentally determined efg's to changes in the donor orbital 
population at nitrogen and, therefore, the extent of electron 
donation to the Lewis acid center. For the imino(N3) nitrogen, 
several general observations can be made. Charge withdrawal 
from N(3) is lower in six-coordinate than in four-coordinate 
metal complexes. Zinc withdraws more charge from nitrogen than 
does cadmium in analogous metal complexes. The extent of charge 
withdrawal in four-coordinate complexes varies with the nature of the 
other ligands bound to the metal. Imidazole and its anion differ 
in their extent of donation to a metal ion [7]. For the amino (Nl) 
nitrogen, the population of the donor orbital, which is involved 
in the N-H bond, is determined primarily by intermolecular inter- 
actions, particularly those involving hydrogen bonding. In 
addition to quantitative information about orbital populations, 
structural information may be obtained [8] . 

The interactions of Zn and Cd with the -NH2 group of amino 
acids and peptides have also been studied using NQR spectroscopy 
[9] . The model employed for sp 3 nitrogen is a modification of 
that employed by Edmonds, et al^ [10] . The N-H bonds are assumed 
equivalent resulting in a plane of symmetry. The populations 
at nitrogen of the orbitals forming the two identical N-H, N-C, 
and third N-H (or N-M) bonds are defined as a, b, and c, re- 
spectively. Upon coordination, the third hydrogen of the zwitterion 
is replaced by the metal ion, changing the population, c. A con- 
comitant change occurs in the other orbitals as given by the 
following equations: 

a = a + A(c - c ) 
o o 

b = b_ + B (c - c ) 
o o 

The values of the scaling factors, A and B, are chosen to be con- 
sistent with those of the previous imidazole study. Values for 
the components of the efg are calculated as a function of the 
population change, (c-c ) . Values of (c-c Q ) for metal-peptide 
and metal-amino acid complexes are derived by fitting the ex- 
perimentally determined efg to the calculated curves. These values 
may be interpreted in terms of the metal-nitrogen interaction. 



67 



References 

1. G. L. Eichhorn, "Inorganic Biochemistry", Elsevier Scientific 
Publishing Company, Amsterdam, 1973. 

2. E. A. C. Lucken, "Nuclear Quadrupole Coupling Constants", 
Academic Press, New York, N.Y., 1969. 

3. C. H. Townes, B. P. Dailey, J. Chem. Phys., 17, 782 (1949). 

4. D. T. Edmonds, Phys. Reports, C. , 29^ 233 (1977). 

5. C. I. H. Ashby, C. P. Cheng, T. L. Brown, J. Am. Chem. Soc, 
100 , 6057 (1978) . 

6. Y. N. Hsieh, G. V. Rubenacker, C. P. Cheng, T. L. Brown, J. 
Am. Chem. Soc, 9^9, 1384 (1977). 

7. C. I. H. Ashby, C. P. Cheng, E. N. Duesler, T. L. Brown, 
J. Am. Chem. Soc, 100 , 6063 (1978). 

8. W. L. McCullen and T. L. Brown, J. Phys. Chem., £1, 2676 (1977) 

9. C. I. H. Ashby and T. L. Brown, manuscript in preparation. 

10. D. T. Edmonds, M. J. Hunt, A. L. Mackay, J. Magn . Reson., 9_, 
66 (1973). 



68 



THE METAL-CARBON TRIPLE BOND 



Allen D. Clauss 



May 17, 1979 



The first compounds containing a metal-carbon triple bond 
were reported by E. 0. Fischer in 1973 [3]. Since that time, 
over one hundred compounds of this type have been characterized. 
The reactivity of these compounds, however, remains relatively 
unexplored [1-12] . 

The first carbyne* compounds synthesized by Fischer and co- 
workers were derived from the well known heteroatom-stabilized 
carbene complexes [7] . They found that treatment of the carbenes 
with boron trihalides resulted in halide substitution at the metal 
center and loss of the lone pair stabilizing group as an anion 
to give thermally unstable carbyne complexes (eq. 1) [3]. 



^R BX 3 
(CO) 5 M=C > X(CO)i»MEC-R 

N 0R 

M = Cr, Mo, W 



(1) 



CH 2 R 



M 



^> 



CHR 



-> MEC-R + H 3 C-R 



(2) 



Alkylidyne compounds of greater thermal stability have been 
prepared by Schrock and coworkers [8-12], In general, these com- 
pounds arise from a-hydrogen abstraction from a sterically crowded, 
mixed alkyl-alkylidene intermediate with concomitant elimination 
of alkane (eq. 2) [8,9,11]. The first examples of this type were 
tantalum and niobium neopentylidyne and benzilidyne complexes [8,9] 
Recently, Schrock has reported neopentylidyne compounds of molyb- 
denum and tungsten by a similar method [12] . 



Although relatively little work has been 
the reactivity of carbyne compounds, recent wo 
Stone has demonstrated interesting and diverse 
for these compounds [13-15] . Kreissl and cowo 
that Fischer-type carbyne complexes of molybde 
can be carbonylated under very mild conditions 
yield of the corresponding metal substituted k 
recently, Stone and coworkers have reported th 
zero-valent platinum center across the metal c 
in a Fischer-type tungsten carbyne to give the 
with a bridging carbyne ligand [15] . 



reported concerning 
rk by Kreissl and 

reaction pathways 
rkers have shown 
num and tungsten 

to give a high 
etene [13] . More 
e addition of a 
arbon triple bond 

mixed metal dimer 



*The terms carbyne and alkylidyne are both used to a considerable 
extent in the chemical literature in reference to compounds con- 
taining a metal carbon triple bond. There is no clear distinction 
between the two terms and they may be used interchangeably. 



69 



References 

1. Ernst Otto Fischer, Carbene and Carbyne Complexes, Adv. 
Organomet. Chem. , \A_, 1 (1976). 

2. E. 0. Fischer and U. Schubert, Ubergangsmetall-Carbin-Komplexe 
VII., J. Organomet. Chem., 100 , 59 (1975). 

3. E. O. Fischer, G. Kreis, C. G. Kreiter, J. Miiller, G. Huttner 
and H. Lorenz, A New Class of Compounds Having a Transition 
Metal-Carbon Triple Bond, Angew. Chem. Int. Ed., 1£, 564 (1973). 

4. E. O. Fischer, G. Huttner, W. Kleine and A. Frank, trans- 
Bromo- (diethylaminocarbyne) tetracarbonylchromium, Angew. Chem. 
Int. Ed., 14, 760 (1975). 

5. E. O. Fischer, G. Huttner, J. Lindner, A. Frank, and F. Kreissl, 
A Transition Metal Carbyne Complex Containing a Metal-Metal 
Bond, Angew. Chem., Int. Ed., 15_, 157 (1976). 

6. E. O. Fischer, M. Schluge and J. O. Besenhard, trans-Bromo- 

(tetracarbonyl) ferrocenylcarbynetungsten , Angew. Chem., Int. 
Ed. , 15, 683 (1976) . 

7. E. O. Fischer and A. Maasbol, Angew. Chem. Int. Ed., 3_, 580 (1964) 

8. L. J. Guggenberger and R. R. Schrock, A Tantalum Carbyne Complex, 
J. Am. Chem. Soc, 9_7, 2935 (1975). 

9. S. J. McLain, C. D. Wood, L. W. Messerle, R. R. Schrock, 
F. J. Hollander, W. J. Youngs and M. R. Churchill, Stable 
Tantalum Alkylidyne Complexes, J. Am. Chem. Soc, 100 , 
5962 (1978) . 

10. J. D. Fellmann, G. A. Rupprecht, C. D. Wood, and R. R. Schrock, 
Bisneopentylidene Complexes of Niobium and Tantalum, J. Am. 
Chem. Soc, 1_0_0, 5964 (1978). 

11. R. R. Schrock and J. D. Fellmann, Mechanism of Formation of 
Neopentylidene Complexes, J. Am. Chem. Soc, 100 , 3359 (1978). 

12. D. N. Clark and R. R. Schrock, Tungsten and Molybdenum 
Neopentylidyne Complexes, J. Am. Chem. Soc, 100 , 6774 (1978). 

13. F. R. Kreissl, W. Uedelhoven, and K. Eberl , Carbonylation of 
Transition-Metal Carbyne Complexes, Angew. Chem. Int. Ed., 17 , 
857 (1978). 

14. F. R. Kreissl, K. Eberl and W. Uedelhoven, Deoxygenation — 

A Novel Reaction Path of Transition Metal-Substituted Ketenes , 
Angew. Chem. Int. Ed., 17, 860 (1978). 

15. T. A. Ashworth, J. A. K. Howard, and F. G. A. Stone, Addition 
of Nucleophilic Metal Complexes to Metal Carbyne Compounds, 
J. C. S. Chem. Comm. , 42 (1979). 



70 



MAGNETIC EXCHANGE INVESTIGATIONS OF TRANSITION METAL COMPLEXES 



Lynn C. Francesconi May 22, 1979 



Polynuclear complexes of Ti(III) and Cu(II) are amenable to 
investigation by EPR and magnetic susceptibility techniques. Such 
studies coupled with molecular orbital analyses can provide impor- 
tant data bearing on the location of unpaired electron density in 
a complex, the facility of electron transfer, and the orbital path- 
ways of electron transfer. 

By virtue of their reactivity, low valent titanium complexes 
have experienced much interest. Recently, bis (n 5 -cyclopentadienyl) - 
dicarbonyltitanium(II) has been found to react with small molecules 
including CO2 , promoting its disproportionation to CO§ and CO [1]. 
Bis (n 5 -cyclopentadienyl) titanium (II) and the zirconium analog re- 
duce dinitrogen to ammonia and hydrazine [2] . TiCli+ and various 
alkoxide modifications are reduced by a trialkyl aluminum to form 
the fibrous Ti(III) active species of the Ziegler-Natta catalyst, 
important in olefin polymerization [3] . A study of the reactivity 
and electronic nature of low valent titanium compounds may contri- 
bute to the overall understanding of the chemistry of this element. 

Complexes of bis (n 5 -cyclopentadienyl) titanium(III) and various 
dicarboxylic acids have been prepared. The compounds do not exhibit 
an appreciable magnetic exchange interaction in the variable tempera- 
ture magnetic susceptibility experiments. EPR studies, however, 
indicate that an exchange interaction is propagated through the 
various bridging units which consist of aliphatic straight chain 
and cyclic dicarboxylic acids and phthalic acids. This is in agree- 
ment with recent EPR evidence that nitroxyl biradicals bridged by 
these moieties experience such spin exchange [4] . Our results cor- 
roborate a recent proposal that a Co (III) binuclear complex, bridged 
by one carboxylate end of a dicarboxylic acid, is reduced by Ti(III), 
which complexes to the free end of the acid and transfers its elec- 
tron through the carbon chain [5] . 

Various pyrimidines and related compounds including sulfur con- 
taining compounds have been successfully complexed to bis (n 5 -cyclo- 
pentadienyl) titanium (III) [6]. The magnetic susceptibility and EPR 
behavior are quite different upon comparison of the oxygen and sul- 
fur analogs [7]. The occurrence of low lying excited states in the 
sulfur compounds and slight structural changes in the bond angles 
and bond distances from the oxygen to sulfur compounds may provide 
the explanation for the unusual behavior. 

Transition metals have been investigated for their role in poly- 
meric linear chains in the quest for new semiconductors. Variable 
temperature magnetic susceptibility experiments have shown that Cu(II), 
Fe(II), and Fe(III) dihydroxyquinone polymers exhibit antif erromagn- 
tism characteristic of linear chains [8]. Because of the uncertainty 
in determining exchange parameters for polymeric compounds, especially 
those those that are not structurally characterized, it is of interest 
to investigate the magnetic exchange present in discrete binuclear 



71 



metal complexes, which have the same bridging units found in the 
polymers . 

Copper (II) and nickel (II) dihydroxyquinone complexes have been 
prepared. EPR and magnetic susceptibility investigations have elu- 
cidated the structure and bonding of these complexes. The viability 
of the various dihydroxyquinone dianion bridging units for propaga- 
ting magnetic exchange interactions has been compared with the 
characteristics of the oxalate and squarate dianions [9] . 



References 

1. G. Fachinetti, C. Floriani, A. Chiesi-Villa , and C. Guastini, 
J. Amer. Chem. Soc . , 1979, 101 , 1767. 

2. J. E. Bercaw, R. H. Marvich, L. G. Bell, and H. H. Brintzinger, 

J. Amer. Chem. Soc , 1972, 9_4, 1219; R. D. Sanner, J. M. Manriquez , 
R. E. Marsh, and J. EZ Bercaw, J. Amer. Chem. Soc , 1976, 9_8, 8351. 

3. M. Tsutsui and A. Courtney, Adv. in Organom. Chem. , 1977, 16_, 241. 

4. K. Mukai and T. Tamaki, Bull. Chem. Soc. Japan , 1977, 50_, 1239; 

E. K. Metzner, L. J. Libertini, and M. Calvin, J Z Amer. Chem. Soc. , 

1977, 99, 4500. 

5. M. Hery and K. Wieghardt, Inorg. Chem. , 1978, 17_, 1130. 

6. B. F. Fieselmann, D. N. Hendrickson, and G. D. Stucky, Inorg. Chem. , 

1978, 17_, 1841; D. R. Corbin, L. C. Francesconi, D. N. Hendrickson, 
ancTG. D. Stucky, J. C. S. Chem. Comm. , 19^79, 248; L. C. Francesconi, 
D. R. Corbin, L. C. Stecher, D. N. Hendrickson, and G. D. Stucky, 
manuscript in preparation. 

7. L. C. Francesconi, D. R. Corbin, D. N. Hendrickson, and G. D. Stucky, 
submitted to Inorg. Chem. 

8. H. Kobayashi, T. Haseda, E. Kanda, and S. Kanda , J. Phys . Soc. Japan , 
19J53, Ij8, 349. J. T. Wrobleski and D. Brown, Inorg. Chem ., 1979, 

18_, 498. 

9. C. G. Pierpont, L. C. Francesconi, and D. N. Hendrickson, Inorg . 
Chem . , 19J77 , 16, 2367; C. G. Pierpont, L. C. Francesconi, and 

D. N. Hendrickson, Inorg. Chem., 1978, 17, 3470. 



i 



72 



BASE CATALYSIS OF LIGAND SUBSTITUTION IN METAL CARBONYLS 

Peter A. Bellus May 24, 1979 

Metal carbonyls are likely to be important species in catalytic 
processes involving carbon monoxide, and therefore their reactions 
are of interest. 

There are two ways in which metal carbonyls can react with 
nucleophiles : 

The first is ligand substitution, which normally proceeds via 
a rate-limiting ligand loss with subsequent rapid attack of the nu- 
cleophile on the coordinatively unsaturated metal center [1,2]. 

The second is attack of a nuclophile at the carbonyl carbon to 
give an adduct of the carbonyl and the nucleophile as a new ligand 
[3,4]. 

Mn(CO)5CH3CN appears to undergo substitution reactions via 
nucleophilic attack at CO. This species is known to undergo nucleo- 
philic attack at CO [5,6]. The substitution mechanism can be un- 
derstood in terms of the altered electronic properties of the new 
ligand, the adduct-ligand, formed by the attack of the nucleophile 
at the CO carbon, relative to CO, which can have a profound effect 
on ligand lability in the metal carbonyl [7] . 

This reaction has been observed in two solvent systems. In ni- 
tromethane it appears that attack of the conjugate base of the solvent 
on coordinated CO occurs. Rapid substitution at the metal follows 
as a result of the labizing power of the adduct-ligand. The product 
is Mn(CO)3L3^ and a mole of CO2 is liberated per mole complex. The 
rate depends on both the nature and concentration of L. 

In solvents where conjugate base formation is unlikely (e.g., 
acetonitrile) it was observed that while triphenylphosphine reacts 
by a dissociative pathway, pyridine reacts much faster and the rate 
depends on pyridine concentration. This can be interpreted in terms 
of direct attack by pyridine at the CO carbon. This adduct-ligand 
labilizes the metal complex towards substitution. 

The concept of adduct-ligands derived from CO imparting labili- 
zation to metal carbonyl complexes has applications to some catalytic 
processes. Both the phase-transfer catalysed substitution reactions 
of group VI metal carbonyls [8] and the iron pentacarbonyl catalyzed 
water-gas shift reaction [9] can be explained using this concept. 



73 



References 

1. F. Basolo, R. G. Pearson, "Mechisms of Inorganic Reactions", 
John Wiley & Sons, N.Y. , 1967, Ch.7. 

2. R. J. Angelici, Organomental . Chem. Rev., 3^, 173 (1968). 

3. R. J. Angelici, Accts. Chem. Res., 5_, 335 (1972). 

4. D. J. Darensbourg, M. Y. Darensbourg, Inorg, Chem., 9_, 1691 (1970). 

5. D. Drew, D. J. Darensbourg, M. Y. Darensbourg, Inorg. Chem., 14 , 
1579 (1975). 

6. D. J. Darensbourg, Isr. J. Chem., 15, 247 (1977). 

7. T. L. Brown, P. A. Bellus, Inorg. Chem., 17^, 3726 (1978). 

8. K. -Y. Hui, B. L. Shaw, J. Organoment, Chem., 124 , 262 (1977). 

9. a. H. -C. Kang, C. H. Mauldin, T. Cole, W. Slegier, K. Cann, 

R. Pettit, J. Am. Chem. Soc, 99, 8323 (1977). 

b. R. M. Laine, R. G. Rinker, P. C. Ford, ibid., 99, 252 (1977). 

c. C. H. Cheng, D. E. Hendriksen, R. Eisenberg, ibid., 99 , 
2791 (1977) . 

d. R. B. Kins, C. C. Frazier, R. M. Hanes , A. D. Kings, Jr., 
ibid. , 100, 2925 (1978) . 



ELECTRON TRANSFER IN MIXED VALENCE METALLOCENES 



74 



Jack A. Kramer 



May 29, 1979 



Electron transfer is an integral part of many chemical, 
electrochemical, and biological processes. In particular, inner 
and outer sphere electron transfer reactions and biological 
electron transport chains involved in metabolism, nitrogen fixation, 
and photosynthesis have received considerable attention. Mixed 
valence compounds [1,2], which contain two or more metal atoms in 
different oxidation states, have been found useful in unravelling 
the intricacies of these electron transfer reactions. Knowledge 
of the factors which influence electron transfer in mixed valence 
compounds should be applicable to understanding these same phen- 
omena in more complicated systems . 

Bridged ferrocenes (I, II) are ideal compounds for the study 
[3-5] of mixed valence interactions because of their structural 
and electronic variability, stability in a variety of oxidation 
states, and well characterized ferrocene and ferricenium electronic 
ground states. They are also amenable to study by such techniques 
as EPR, IR, NMR, electronic, and 57 Fe Mossbauer spectroscopies, and 
variable temperatures magnetic susceptibility. 




Fe 




Fe 



Fe 



Fe 



ii 



a. 
b. 



X = 
X = 



-CEC- 
-Se- 



a. X = -CEC- 



By varying the structure of the bridged ferrocene unit and 
determining the rate of electron transfer by the use of various 
physical techniques, information pertaining to the different re- 
quirements of the direct and superexchange electron transfer 
mechanisms may be obtained. 



The unpaired electron of the bisfulvalene dicobalt monocation 
[6] is directed toward the bridging ligand, while that of the 
bisfulvalene diiron monocation (Class III, delocalized) is directed 
toward the other iron. If the rate of electron transfer were much 



75 



less in the cobalt case, then the direct mechanism of electron 
transfer must be dominant. However, optical and EPR studies show 
that the electron of the cobalt compound is also delocalized. 

The mixed valence cation of diferrocenyl selenide [7] (lb) 
was found to be a Class I (noninteracting) mixed valence compound. 
No intervalence transfer (IT) band was seen in the electronic 
spectrum, and the 7 Fe Mossbauer spectrum [8] exhibits separate 
absorptions due to the ferrocene and ferricenium portions of the 
molecule. An Fe-Fe distance of 6.06 & was derived from the X-ray 
structure of DFSI 3 • 1 2 • %CH 2 Cl2 . Comparison of this distance with 
those of ferricenyltris (ferrocenyl) borate [9], a Class II (weakly 
interacting) mixed valence compound, leads to the conclusion that 
the maximum distance for appreciable direct electron transfer is 
between 5.35 and 6.06 A. 

In an effort to establish a similar limit for superexchange, 
diferrocenylacetylene was prepared and oxidized. Although the 
mixed valence cation could be generated electrochemically [10] , 
chemical oxidation produced only the dioxidized species. Variable 
temperature magnetic susceptibility measurements showed that there 
is a 3 cm -1 interaction between the two Fe(III) ions. This comple- 
ments the finding that the monocation is weakly interacting [10,11] 

[2 . 2] ferrocenophane-1 , 13-diyne (Ila) was prepared to attempt 
to increase the amount of interaction, analogous to the biferrocene- 
bisfulvalene diiron case. The mixed valence ion was found to have 
equivalent iron sites on the timescales of 57 Fe Mossbauer, EPR, 
and IR spectroscopies [10-12] . Because of the long Fe-Fe distance 
(=6.5 A) in this compound, superexchange must be responsible for 
this large interaction. In order for this superexchange to occur, 
the bridged cyclopentadienide groups must be coplanar for maximum 
overlap of the pi systems. 

A correlation was found between the quadrupole splittings of 
the 7 Fe Mossbauer spectra of various oxidized mono- and binuclear 
ferrocene compounds and the g anisotropics observed in the EPR 
spectra. It is proposed [12] that the orbital angular momentum 
of the unparied electron is being quenched by distortion and/or 
delocalization, with a corresponding redistribution of electrons 
among the Fe 3d orbitals , which is reflected by the quadrupole 
splitting. 

References 

1. M. B. Robin and P. Day, Adv. Inorg. Chem. Radiochem ., 10 , 247 

(1967) . 

2. G. C. Allen and N. S. Hush, Prog. Inorg. Chem ., 8_, 257 (1967). 

3. D. 0. Cowan, C. Levanda, J. Park, and F. Kaufman, Ace. Chem . 
Res. ,6,1 (1973) . 



76 

4. W. H. Morrison, Jr. and D. N. Hendrickson, Inorg. Chem . , 14, 
2331 (1975). 

5. M. J. Powers and T. J. Meyer, J. Amer. Chem. Soc , 100, 
4393 (1978) . 

6. J. C. Smart, Ph.D. Thesis, Massachusetts Institute of Technology 

(1974) . 

7. P. Shu, K. Bechgaard, and D. 0. Cowan, J. Org. Chem . , 41 , 
1849 (1976) . 

8. J. A. Kramer, F. H. Herbstein, and D. N. Hendrickson, manu- 
script in preparation. 

9. D. 0. Cowan, P. Shu, F. L. Hedberg, M. Rossi, and T. J. 
Kistenmacher, J. Amer. Chem. Soc , 101 , 1304 (1979). 

10. C. LeVanda, K. Bechgaard, and D. 0. Cowan, J . Org . Chem . , 
41, 2700 (1976) . 

11. I. Motoyama, M. Watanabe, and H. Sano, Chem. Lett . , 1978 , 513. 

12. J. A. Kramer and D. N. Hendrickson, manuscript in preparation. 



I 



VMJ